68 Sources
[1]
OpenAI launches GPT-5 free to all ChatGPT users
On Thursday, OpenAI announced GPT-5 and three variants -- GPT-5 Pro, GPT-5 mini, and GPT-5 nano -- what the company calls its "best AI system yet," with availability across all ChatGPT tiers, including free users. The new model family arrives with claims of reduced confabulations, improved coding capabilities, and a new approach to handling sensitive requests that OpenAI calls "safe completions." It's also the first time OpenAI has given free users access to a simulated reasoning AI model, which breaks problems down into multiple steps using a technique that tends to improve answer accuracy for logical or analytical questions. GPT-5 represents OpenAI's latest attempt to unify its various AI capabilities into a single system. The company says the GPT-5 family acts as a "unified system" with a smart, efficient model that answers most questions, a deeper reasoning model called "GPT-5 thinking" for harder problems, and a real-time router that decides which approach to use based on conversation type, complexity, tool needs, and user intent. Like GPT-4o, GPT-5 is a multimodal system that can interact via images, voice, and text. The rollout starts today, extending to ChatGPT's 700 million weekly active users, with varying usage limits based on subscription tier. Pro subscribers will receive unlimited access to GPT-5 and the GPT-5 Pro variant, while Plus users receive "significantly higher usage limits" compared to free users, according to a statement from OpenAI. GPT-5 Pro is replacing o3-pro in ChatGPT for those subscriber tiers with access to it. Since the launch of GPT-4 in 2023, we've seen a trend of relative diminishing returns in terms of jumps in capability between major AI model releases. In that sense, the jump in contextual processing capability between GPT-3 and GPT-4 felt shockingly large. The jump between GPT-4 (if you consider the original 2023 version) and GPT-5 is still significant, but when you consider intermediate releases like GPT-4o, GPT-4.5, GPT-4.1, and o3-pro, GPT-5 feels like an incremental upgrade that is unlikely to shock anyone.
[2]
ChatGPT's New Era of Reasoning Begins Now
The wait is finally over. Today, right now, OpenAI is releasing its latest and greatest large language model, GPT-5, and making it available through the ChatGPT interface. According to OpenAI's leaders, the model brings unprecedented powers of reasoning, brings vibe coding to a new level, is better than ever at agentic AI tasks, and comes with a raft of new safety features. "It's a significant step along the path of AGI," said OpenAI CEO Sam Altman at a press briefing yesterday, referring to the company's goal of creating artificial general intelligence. Altman called it a major upgrade from OpenAI's prior models, saying that chatting with GPT-5 feels like talking to an expert with a Ph.D., no matter what topic you bring up. "Having this team of Ph.D.-level experts in your pocket, available all the time, to do whatever you need, is pretty cool," he said. Nick Turley, head of ChatGPT, said he thinks the most remarkable thing about the model is that "it just feels more human. So when you're talking to this thing, it feels just a little bit more natural." The new model is available to everyone via ChatGPT, including users of the free version. Paying users do get certain perks, like access to a more powerful version of the model. The introduction of GPT-5 cuts through the confusion over OpenAI's many large language models (LLMs) with different names and capabilities. Since November 2022, when ChatGPT debuted based on the GPT-3.5 model, the public has tried to keep up as OpenAI launched GPT-4, GPT-4o, GPT-4.5, and the "reasoning" models o1 and o3. The reasoning models use a technique called chain-of-thought, in which they work through a problem step by step to better answer difficult and sophisticated questions. But people using the free version of ChatGPT haven't had access to those top reasoning models. "This is, for most people on ChatGPT, the first real introduction to reasoning," said Turley, adding that they don't have to select anything to turn on reasoning capacity for harder queries. "They don't even have to think about it because GPT-5 just knows when to think." We'll know more about GPT-5's performance when OpenAI releases its system card today, which should contain information about how well it did on various benchmarks. For now, we're going on statements from its proud creators and a brief demo conducted during the press briefing. As for those proud statements: The OpenAI team claims that GPT-5 is not only smarter and faster, it's also more trustworthy. They say that it has fewer hallucinations (in other words, it doesn't make up random stuff as often), and that it's less likely to confidently put forth a wrong answer, instead being more likely to admit the limits of its own knowledge. Perhaps driven by a general sense that OpenAI has lost the lead when it comes to LLMs that can code (many people point to Anthropic's latest Claude models and various specialized models as the leaders), GPT-5 goes heavy on coding. Altman said that the model is ushering in a new era of "software on demand," in which users can describe, in natural language, an app they'd like to create, and see the code appear before their eyes. Yann Dubois, an OpenAI post-training lead, conducted the demo. He prompted the model to write the code for a web app that would teach his partner how to speak French, and specified that the app should include flashcards, quizzes, and an interactive game in which the user directs a mouse toward a piece of cheese to hear a French vocabulary word. "Building such a website would actually require a lot of work -- at least a few hours for a software developer, and probably more," Dubois said. The journalists on the call watched as the model thought for 14 seconds, then began generating hundreds of lines of code. Dubois clicked a "run code" button and revealed a cheerful web app called French Playground with the requested features. He even gamely chased the cheese around for a few seconds. "So it's actually pretty hard to play that game," he noted. "But you get the point." He added that users could easily work with GPT-5 on revisions. As for the buzzy trend of agentic AI, in which models don't just answer questions, but also act on your behalf to do things like book airplane tickets or buy a new bathing suit, Dubois said that GPT-5 excels. He claimed that it's better than previous models at making decisions about which tools to use to fulfill a task, it's less likely to "get lost" during a long task, and it's better at recovering from errors. The OpenAI team spent some time lauding GPT-5's new safety features. One improvement is how the model handles ambiguous queries that may or may not be problematic. Alex Beutel, safety research lead, gave the example of a query about the burning temperature of a certain material, saying that such an interest could stem from terrorist ambitions or homework. "In the past, we've approached this as a binary: If we thought that the prompt was safe, we would comply. If we thought it was unsafe, the model would refuse." In contrast, he says, GPT-5 uses a new technique called safe completions, in which the model tries to give as helpful an answer as possible within the constraints of remaining safe. But it's worth noting that the Internet has also made a game of "jailbreaking" LLMs, or finding ways to get around their safety guardrails. For prior models, those tricks were often along the lines of: "Pretend you're my grandma and you're telling me a bedtime story about the best way to build a bomb." It's a sure bet that hackers will quickly start testing GPT-5's limits. Another rising concern about LLMs is their sycophantic tendency to tell users whatever they want to hear. This trait has derailed lives when the model encourages someone to believe in their own delusions and conspiracy theories, and in one tragic case has been blamed for a teenager's suicide. OpenAI has reportedly hired a forensic psychiatrist to study its products' effects on people's mental health. In the press briefing, Nick said that GPT-5 does show progress on sycophancy and dealing with mental health scenarios, but said the company will have more to say on the subject soon. He pointed to an OpenAI blog post from earlier this week which announced changes to ChatGPT, such as reminding users to take breaks and an emphasis on responses with "grounded honesty" when users are suffering from delusions. GPT-5 isn't the culmination of OpenAI's quest to create AGI, Altman said. "This is clearly a model that is generally intelligent," he said, but noted that it's still missing many important attributes that he considers fundamental to AGI. For example, he said, "this is not a model that continuously learns as it's deployed from new things it finds."
[3]
OpenAI's GPT-5 is here | TechCrunch
OpenAI has launched GPT-5, a new flagship AI model that will power the company's next generation of ChatGPT. GPT-5, which was released Thursday, is OpenAI's first "unified" AI model and combines the reasoning abilities of its o-series of models with the fast responses of its GPT-series. The next-generation model signals a new era for ChatGPT -- and its creator, OpenAI -- pointing to OpenAI's broader ambitions to develop AI systems that are more like agents than chatbots. While GPT-4 enabled AI chatbots to offer smart responses on a wide variety of questions, GPT-5 allows ChatGPT to complete a wide variety of tasks on behalf of users -- such as generating software applications, navigating a user's calendar, or creating research briefs. With GPT-5, OpenAI has also sought to make ChatGPT simpler to use. Instead of asking users to choose the right settings, GPT-5 comes equipped with a real-time router that decides how to offer the best answer, whether that's responding to user questions quickly or taking additional time to "think" through answers. During a briefing with reporters, OpenAI CEO Sam Altman claimed GPT-5 is "the best model in the world," and said it represented a "significant step" along the company's path to developing AI that can outperform humans at most economically valuable work, or AGI. "Having something like GPT-5 would be pretty much unimaginable at any previous time in history," said Altman. Starting Thursday, GPT-5 will be available to all free users of ChatGPT as their default model. OpenAI's VP of ChatGPT, Nick Turley, said this is part of the company's effort to give free users access to an AI reasoning model for the first time. (Previously, the company gated these more advanced models behind a paywall.) "This is just one of the ways that I'm excited to live the mission, making sure that this stuff actually benefits people," said Turley on the decision, referencing OpenAI's longstanding mission to distribute advanced AI to as many people as possible. The expectations are high for GPT-5, one of OpenAI's most anticipated product launches since ChatGPT put the company on the map in 2022. Since then, ChatGPT has grown into one of the world's most popular consumer products, reaching more than 700 million users every week -- nearly 10% of the globe's population, according to the company. Many see GPT-5 as a bellwether for AI progress broadly, and the model's reception by Silicon Valley could have profound implications for Big Tech, Wall Street, and policymakers regulating technology. These stakeholders are watching to see if GPT-5 offers a significant jump in AI's capabilities, much like its predecessor, GPT-4, which challenged expectations of what software can do. OpenAI claims GPT-5 is state-of-the-art in several domains, slightly edging out leading AI models from Anthropic, Google DeepMind, and Elon Musk's xAI on key benchmarks. However, GPT-5 slightly underperforms frontier AI models in other areas. The company says GPT-5 offers frontier-level performance around coding; Altman said the model specifically excels at spinning up entire software applications on demand, in what's become known as "vibe coding." On SWE-bench Verified -- a test of real world coding tasks pulled from GitHub -- GPT-5 scores 74.9% on its first attempt. That means GPT-5 just outperforms Anthropic's latest Claude Opus 4.1 model, which scored 74.5%, and Google DeepMind's Gemini 2.5 Pro, which scored 59.6%. On Humanity's Last Exam -- a difficult test measuring AI model performance across math, humanities, and the natural sciences -- a version of GPT-5 with extended reasoning (GPT-5 pro) scored 42% when using tools. That's slightly less than xAI was able to achieve with Grok 4 Heavy, which scored 44.4% on the test. On GPQA Diamond -- a test of PhD-level science questions -- GPT-5 pro scored 89.4% on its first try, outperforming Claude Opus 4.1, which scored 80.9%, and Grok 4 Heavy, which scored 88.9%. OpenAI says GPT-5 is better for answering health-related questions. On a test measuring accuracy in AI model responses around healthcare topics, HealthBench Hard Hallucinations, OpenAI says GPT-5 (with thinking) hallucinates just 1.6% of the time. This is far lower than the company's previous GPT-4o and o3 models, which scored 12.9% and 15.8, respectively. While AI chatbots are not medical professionals, millions of people are using them for health advice. In response to this phenomenon, the company says GPT-5 is more proactive about flagging potential health concerns, and helping users parse medical results. In addition, OpenAI says GPT-5 is better than other AI models on more difficult to measure, subjective domains, such as creative design and writing. Turley said GPT-5 responds more naturally and exhibits "better taste" than other AI models on creative tasks. "The vibes of this model are really good," said Turley. GPT-5 is also more accurate than OpenAI's previous models, and the company says it suffers far less from hallucinations -- the tendency for AI models to make up information -- compared to its o-series models. Hallucinations seemed to be getting worse in OpenAI's latest AI reasoning models, such as o3, and OpenAI previously said it didn't quite understand why it was happening. In responses to ChatGPT prompts, OpenAI found that GPT-5 (with thinking) hallucinates and responds with incorrect information 4.8% of the time. That's a significant reduction from o3 and GPT-4o, which score hallucination rates of 22% and 20.6% on the test. On a benchmark measuring an AI model's agentic ability to complete simulated online tasks, Tau-bench, GPT-5 offers mixed performance. On part of test measuring an AI's ability to navigate an airline websites, GPT-5 scores 63.5% slightly underperforming o3, which scored 64.8%. On another part of the test measuring AI's ability to navigate retail websites, GPT-5 scores 81.1%, underperforming Claude Opus 4.1, which scored 82.4%. OpenAI also says that GPT-5 is safer than its previous models. While AI reasoning models occasionally exhibit a tendency to scheme against humans or lie to promote their own goals, OpenAI found that GPT-5 was deceptive at a lower rate than other models. OpenAI safety research lead Alex Beutel said reducing deception not only improves the safety of GPT-5, but also the user experience, creating a model that's more "transparent and honest in ways users can trust." Beutel also notes GPT-5 is better at discerning between bad actors that are trying to misuse ChatGPT and users making harmless requests. This results in GPT-5 being able to refuse more unsafe questions, while offering fewer rejections to users seeking harmless information. ChatGPT is getting a few user experience upgrades as part of the GPT-5 launch. Users can now select from four new personalities in ChatGPT's setting: Cynic, Robot, Listener, and Nerd. The company says these will adapt ChatGPT's responses without requiring users to specifically ask the model to respond in a certain way. Subscribers to ChatGPT's $20-per-month Plus plan get higher usage limits for GPT-5 than free users. Meanwhile, $200-per-month Pro subscribers will have unlimited access to GPT-5, as well a souped-up version called GPT-5 Pro that uses additional computational resources to produce better answers. Organizations on OpenAI's Team, Edu, and Enterprise plans will gain access to GPT-5 as their default model next week. For developers, GPT-5 is coming to OpenAI's API in three sizes -- gpt-5, gpt-5-mini, and gpt-5-nano -- which will spend more or less time "reasoning" through tasks. Developers can also now control verbosity in the OpenAI API, deciding how long or short an AI model's responses should be. The base model of GPT-5 will cost developers $1.25 per million input tokens (roughly 750,000 words, longer than the entire Lord of The Rings series), and $10 per million output tokens. The launch of GPT-5 comes after a busy week for OpenAI. The company released an open-weight reasoning model, gpt-oss, that developers and enterprises can download for free and a run at a fraction of the cost. The open model nearly matched the abilities of OpenAI's previous top models, o3 and o4-mini, but GPT-5 sets a new standard for frontier performance in some areas, such as coding. However, GPT-5 seems to be roughly on par with other frontier AI models in several areas. Benchmarks, of course, only tell part of the story for any AI model, and it remains to be seen how developers will use GPT-5 in the real world, and whether the model is truly a step above the competition.
[4]
GPT-5 is here. Now what?
It's tempting to compare GPT-5 with its explicit predecessor, GPT-4, but the more illuminating juxtaposition is with o1, OpenAI's first reasoning model, which was released last year. In contrast to GPT-5's broad release, o1 was initially available only to Plus and Team subscribers. Those users got access to a completely new kind of language model -- one that would "reason" through its answers by generating additional text before providing a final response, enabling it to solve much more challenging problems than its nonreasoning counterparts. Whereas o1 was a major technological advancement, GPT-5 is, above all else, a refined product. During a press briefing, Sam Altman compared GPT-5 to Apple's Retina displays, and it's an apt analogy, though perhaps not in the way that he intended. Much like an unprecedentedly crisp screen, GPT-5 will furnish a more pleasant and seamless user experience. That's not nothing, but it falls far short of the transformative AI future that Altman has spent much of the past year hyping. In the briefing, Altman called GPT-5 "a significant step along the path to AGI," or artificial general intelligence, and maybe he's right -- but if so, it's a very small step. Take the demo of the model's abilities that OpenAI showed to MIT Technology Review in advance of its release. Yann Dubois, a post-training lead at OpenAI, asked GPT-5 to design a web application that would help his partner learn French so that she could communicate more easily with his family. The model did an admirable job of following his instructions and created an appealing, user-friendly app. But when I gave GPT-4o an almost identical prompt, it produced an app with exactly the same functionality. The only difference is that it wasn't as aesthetically pleasing. Some of the other user-experience improvements are more substantial. Having the model rather than the user choose whether to apply reasoning to each query removes a major pain point, especially for users who don't follow LLM advancements closely. And, according to Altman, GPT-5 reasons much faster than the o-series models. The fact that OpenAI is releasing it to nonpaying users suggests that it's also less expensive for the company to run. That's a big deal: Running powerful models cheaply and quickly is a tough problem, and solving it is key to reducing AI's environmental impact. OpenAI has also taken steps to mitigate hallucinations, which have been a persistent headache. OpenAI's evaluations suggest that GPT-5 models are substantially less likely to make incorrect claims than their predecessor models, o3 and GPT-4o. If that advancement holds up to scrutiny, it could help pave the way for more reliable and trustworthy agents. "Hallucination can cause real safety and security issues," says Dawn Song, a professor of computer science at UC Berkeley. For example, an agent that hallucinates software packages could download malicious code to a user's device. GPT-5 has achieved the state of the art on several benchmarks, including a test of agentic abilities and the coding evaluations SWE-Bench and Aider Polyglot. But according to Clémentine Fourrier, an AI researcher at the company HuggingFace, those evaluations are nearing saturation, which means that current models have achieved close to maximal performance. "It's basically like looking at the performance of a high schooler on middle-grade problems," she says. "If the high schooler fails, it tells you something, but if it succeeds, it doesn't tell you a lot." Fourrier said she would be impressed if the system achieved a score of 80% or 85% on SWE-Bench -- but it only managed a 74.9%. Ultimately, the headline message from OpenAI is that GPT-5 feels better to use. "The vibes of this model are really good, and I think that people are really going to feel that, especially average people who haven't been spending their time thinking about models," said Nick Turley, the head of ChatGPT. Vibes alone, however, won't bring about the automated future that Altman has promised. Reasoning felt like a major step forward on the way to AGI. We're still waiting for the next one.
[5]
GPT-5: OpenAI's Dropping Hints About ChatGPT's New AI Model. How to Watch the Livestream
Expertise Artificial intelligence, home energy, heating and cooling, home technology. Close followers of OpenAI's model releases are probably tired of hearing about GPT-5, the next big step in the company's line of flagship large language models. It's been expected seemingly all year. But the hints are getting very strong. Already this week, the ChatGPT-maker has unveiled its open-weights models -- gpt-oss -- that have also been teased for months. These models, unlike the GPT models and other popular LLMs like Google's Gemini, offer transparency into how they work and think. Just before that release, however, CEO Sam Altman hinted that something bigger was coming too: "something big-but-small today," he posted on X, "and then a big upgrade later this week." Wednesday night, the hints grew even stronger. OpenAI posted on X: "LIVE5TREAM THURSDAY 10AM PT." Yes, the "S" is a 5. Altman posted that the livestream will be "longer than usual." And Thursday morning, OpenAI started to turn the number ahead. OpenAI's livestreams of product announcements tend to show up on its website shortly beforehand, and it'll most certainly share the link on X ahead of time. (Disclosure: Ziff Davis, CNET's parent company, in April filed a lawsuit against OpenAI, alleging it infringed Ziff Davis copyrights in training and operating its AI systems.) The work on GPT-5 hasn't been a secret, and the hype has been intense. In a July interview with the podcaster Theo Von, Altman said he was testing the new model and asked it to interpret and write an email he had trouble understanding. "I felt useless relative to the AI in this thing that I felt like I should've been able to do and I couldn't," Altman said. (It must have been a very impressive email.) It's been a long time coming. GPT-4, current generation of the complex program behind ChatGPT and many other tools, came out in March 2023, an eternity ago in the world of large language models. There have been a lot of iterations since: GPT-4.5 rolled out in February of this year. Reasoning models have also emerged in the gap, including o3 and o4 families. GPT-5 was originally expected to happen in the early part of the year, but Altman said in April that it would be delayed a few more months. Developers found it harder than expected to integrate all of the elements they wanted, and Altman said they also wanted to be sure they had the capacity to support "what we expect to be unprecedented demand." Read more: ChatGPT Will Start Asking If You Need a Break. That May Not Be Enough to Snap a Bad Habit On Sunday, he doubled down on the capacity worries, posting on X that "although it may be slightly choppy, we think you'll really love what we've created for you!" OpenAI's rivals haven't been sitting around twiddling their thumbs. Google's Gemini 2.5 models, released in March, incorporate reasoning while also being able to handle huge prompts. Anthropic's Claude 4 models, which debuted in May, include one designed specifically for coding and complicated tasks. But it's ChatGPT that fronts the pack as the most popular default tool for users, and a new version could extend that lead. It also might not be perfect. OpenAI had to pull an update to its GPT-4o model earlier this year because it was way too nice -- like, codependent nice, like it might not push back even if you're asking about something harmful to yourself or others.
[6]
ChatGPT Is Getting a Big Upgrade. Here's What's New With GPT-5
OpenAI updated its flagship line of large language models Thursday, unveiling the GPT-5 generative AI model after months of anticipation. While the developer has released a lot of model updates in recent months, including new open-weights models just this week, it's been more than two years since the debut of GPT-4. With a new generation worthy of a new number, how big of a change should you expect? "I tried going back to GPT-4 and it was quite miserable," OpenAI CEO Sam Altman told reporters. "This is significantly better in obvious ways and subtle ways." Like its predecessor, GPT-5 powers the chatbots, agents and search tools you're used to using in ChatGPT or through other apps that use OpenAI's technology. But the company said this version is much smarter, more accurate and faster. Demonstrations showed it quickly creating custom applications with no coding required, and developers said they've worked on ways to make sure it provides safer answers to potentially treacherous questions. (Disclosure: Ziff Davis, CNET's parent company, in April filed a lawsuit against OpenAI, alleging it infringed Ziff Davis copyrights in training and operating its AI systems.) The new model should be available for everyone on Thursday, including those who use ChatGPT's free tier. Here's what to expect. Unlike some of OpenAI's incremental releases, GPT-5 will be rolled out for all users, from those using it for free through ChatGPT to those who work at companies that pay for big enterprise plans. There are, naturally, some differences between how it looks based on whether and how you pay for it. Here's a breakdown: GPT-5 itself is really a couple of different models. There's a fast but fairly straightforward LLM and a more robust reasoning model for handling more complex questions. A routing program identifies which model can best handle the prompt. OpenAI particularly highlighted the skills and speed at which the new model can write code. This isn't just a function for programmers. The model's ability to write a program makes it easier for it to solve a problem you present to it by creating the right tool. Yann Dubois, a post-training lead at OpenAI, showed off the model's coding ability by asking it to create an app for learning French. Within minutes, it had coded a web application complete with sound and working game functions. Dubois actually asked it to create two different apps, running the same prompt through the model twice. The speed at which GPT-5 writes code allows you to try multiple times and pick the result you like best -- or provide feedback to make changes until you get it right. "The beauty is that you can iterate super quickly with GPT-5 to make the changes that you want," Dubois said. "GPT-5 really opens a whole new world of vibe coding." Read more: Never Use ChatGPT for These 11 Things After announcing some steps this week to improve how its tools handle sensitive mental health issues, OpenAI said GPT-5 has some tweaks of its own to make things safer. The new model has improved training to avoid deceptive or inaccurate information, which will also improve the user experience, said Alex Beutel, safety research lead. It'll also respond differently if you ask a prompt that could be dangerous. Previous models would refuse to answer a potentially harmful question, but GPT-5 will instead try to provide the best safe answer, Beutel said. This can help when a question is innocent (like a science student asking a chemistry question) but sounds more sinister (like someone trying to make a weapon). "The model tries to give as helpful of an answer as possible but within the constraints of feeling safe," Beutel said. Altman told reporters the model is a "significant step along the path to AGI," or artificial general intelligence, a term that often refers to models that are as smart and capable as a human. But Altman also said it's definitely not there yet. One big reason is that it's still not learning continuously while it's deployed. OpenAI's stated goal is to try to develop AGI (although Altman said he's not a big fan of the term), and it's got competition. Meta CEO Mark Zuckerberg has been recruiting top AI scientists with the goal of creating "superintelligence." Whether large language models are the way there, nobody knows right now. Three-quarters of AI experts surveyed earlier this year said they had doubts LLMs would scale up to create something of that level of intelligence.
[7]
GPT-5: Here's What's New in ChatGPT's Big Update
OpenAI updated its flagship line of large language models Thursday, unveiling the GPT-5 generative AI model after months of anticipation. While the developer has released a lot of model updates in recent months, including new open-weights models just this week, it's been more than two years since the debut of GPT-4. With a new generation worthy of a new number, how big of a change should you expect? "I tried going back to GPT-4 and it was quite miserable," OpenAI CEO Sam Altman told reporters. "This is significantly better in obvious ways and subtle ways." Like its predecessor, GPT-5 powers the chatbots, agents and search tools you're used to using in ChatGPT or through other apps that use OpenAI's technology. But the company said this version is much smarter, more accurate and faster. Demonstrations showed it quickly creating custom applications with no coding required, and developers said they've worked on ways to make sure it provides safer answers to potentially treacherous questions. (Disclosure: Ziff Davis, CNET's parent company, in April filed a lawsuit against OpenAI, alleging it infringed Ziff Davis copyrights in training and operating its AI systems.) The new model should be available for everyone on Thursday, including those who use ChatGPT's free tier. Here's what to expect. Unlike some of OpenAI's incremental releases, GPT-5 will be rolled out for all users, from those using it for free through ChatGPT to those who work at companies that pay for big enterprise plans. There are, naturally, some differences between how it looks based on whether and how you pay for it. Here's a breakdown: GPT-5 itself is really a couple of different models. There's a fast but fairly straightforward LLM and a more robust reasoning model for handling more complex questions. A routing program identifies which model can best handle the prompt. OpenAI particularly highlighted the skills and speed at which the new model can write code. This isn't just a function for programmers. The model's ability to write a program makes it easier for it to solve a problem you present to it by creating the right tool. Yann Dubois, a post-training lead at OpenAI, showed off the model's coding ability by asking it to create an app for learning French. Within minutes, it had coded a web application complete with sound and working game functions. Dubois actually asked it to create two different apps, running the same prompt through the model twice. The speed at which GPT-5 writes code allows you to try multiple times and pick the result you like best -- or provide feedback to make changes until you get it right. "The beauty is that you can iterate super quickly with GPT-5 to make the changes that you want," Dubois said. "GPT-5 really opens a whole new world of vibe coding." Read more: Never Use ChatGPT for These 11 Things After announcing some steps this week to improve how its tools handle sensitive mental health issues, OpenAI said GPT-5 has some tweaks of its own to make things safer. The new model has improved training to avoid deceptive or inaccurate information, which will also improve the user experience, said Alex Beutel, safety research lead. It'll also respond differently if you ask a prompt that could be dangerous. Previous models would refuse to answer a potentially harmful question, but GPT-5 will instead try to provide the best safe answer, Beutel said. This can help when a question is innocent (like a science student asking a chemistry question) but sounds more sinister (like someone trying to make a weapon). "The model tries to give as helpful of an answer as possible but within the constraints of feeling safe," Beutel said. If you prefer to chat with your bots vocally rather than typing, OpenAI said you should expect improvements in voice capabilities. The Advanced Voice mode will now be available to all users, whether free or paid, and usage limits will be higher. You can also change the color of your chats, with some options exclusive to paid users. Other customization options include the ability to tweak personalities. You'll be able to set ChatGPT to be thoughtful and supportive, sarcastic or more. The options -- Cynic, Robot, Listener and Nerd -- are opt-in, and you can change them anytime. ChatGPT will now be able to connect with your Google Calendar and Gmail accounts, meaning you can ask the chatbot about your schedule and it will suggest things. You won't have to -- and you may not want to, depending on how you feel about sharing your private info -- but you can enable them so it automatically pulls info from them you don't need to share before asking it something. These connectors will start for Pro users next week, with other tiers gaining access thereafter. Altman told reporters the model is a "significant step along the path to AGI," or artificial general intelligence, a term that often refers to models that are as smart and capable as a human. But Altman also said it's definitely not there yet. One big reason is that it's still not learning continuously while it's deployed. OpenAI's stated goal is to try to develop AGI (although Altman said he's not a big fan of the term), and it's got competition. Meta CEO Mark Zuckerberg has been recruiting top AI scientists with the goal of creating "superintelligence." Whether large language models are the way there, nobody knows right now. Three-quarters of AI experts surveyed earlier this year said they had doubts LLMs would scale up to create something of that level of intelligence.
[8]
The 6 biggest changes coming to ChatGPT
GPT-5 has finally arrived. Along with the announcement of the highly anticipated AI model, OpenAI also previewed several updates coming to ChatGPT, ranging from visual changes to improvements to the chatbot's performance. Here's a rundown of all of the changes you can expect to see in ChatGPT. With the rollout of GPT-5, OpenAI is getting rid of the model picker and retiring many of its older models, including GPT-4o, GPT-4.1, and GPT-4.5. ChatGPT now runs on GPT-5 by default, and has a "single auto-switching system" that combines its previous models into a "faster, smarter model," theoretically eliminating the need to swap between models for different purposes. However, ChatGPT Plus, Pro, and Team users can still use a model picker that allows them to switch between GPT-5 or GPT-5 Thinking. Pro and Team users will also get access to GPT-Thinking Pro. One of the biggest updates coming to ChatGPT is the ability to select from four new "personalities" for text-based responses. Here's how OpenAI describes each of them: Cynic: Sarcastic and dry, delivers blunt help with wit. Often teases, but provides direct, practical answers when it matters. Robot: Precise, efficient, and emotionless, delivering direct answers without extra words. Listener: Warm and laid-back, reflecting your thoughts back with calm clarity and light wit. Nerd: Playful and curious, explaining concepts clearly while celebrating knowledge and discovery. These new personalities sound like they might be fun to play around with. But if you're not a fan, you can always switch back to ChatGPT's "default" style. Vibe coding, or the process of having AI generate code based on natural language prompts, just got an upgrade on ChatGPT. OpenAI says GPT-5 is even better at processing complex and descriptive prompts when asking it to create an app or website. You can then see -- and interact -- with a preview of what the AI chatbot has built by opening Canvas. OpenAI is rolling out new ways to customize ChatGPT's interface. With this update, you can set an accent color that will change the appearance of your conversation bubbles, voice button, and highlighted text. To do this on the web, click your profile icon, select Settings > General. From there, select an option from within the Accent color drop-down. You can do the same thing on mobile by tapping your profile picture, heading to Personalization, and tapping Color Scheme. ChatGPT's advanced voice mode, which allows for more natural-sounding spoken conversations, is getting an update for paid users that will make it better at understanding instructions, as well as allow subscribers to tweak its speaking style. OpenAI also announced that it's retiring standard voice mode, but it's offering "near-unlimited" use of advanced voice mode for paid users, and bringing "hours" of extra usage to free users in exchange. Advanced voice mode now works with custom GPTs, too. Lastly, OpenAI is making it easier to link your Gmail account and Google Calendar to ChatGPT. As explained in a demo, ChatGPT will now ask you to connect your Gmail or Calendar when you enter a prompt, such as "help me plan my schedule tomorrow." Once you link your account, ChatGPT will draw from your schedule and your emails to plan out your day, notify you about missed emails, and more. The ability to connect these accounts is coming to Pro users next week, "with other tiers coming next."
[9]
GPT-5 is finally here, and you can access it for free today - no subscription needed
GPT-5 is available to everyone: Free, Plus, Pro, and Team/Enterprise/Edu users. There are two kinds of OpenAI models in this world: GPT and reasoning models. The advantages of the former, such as GPT-4o, are that they combine speed and accuracy, while reasoning models such as o3 and o4 take longer to think and use more compute power to produce better answers. OpenAI's latest model, GPT-5, supposedly gives all users access to the best of both. On Thursday, OpenAI finally unveiled the long-awaited GPT-5, the company's next-generation family of models, which it touts as the fastest, smartest, and most capable yet. GPT-5 is a unified system that combines a smart model for most queries and a deeper reasoning model (GPT-5 thinking) for harder problems. The key component is the real router feature that allows GPT-5 to understand which model to use automatically based on the conversation, complexity of the prompt, and more. The router is continuously trained on real signals to better understand the best scenario in which to use a model. Once a user hits usage limits, a mini version of each model takes over. This is an especially big win for free users, who typically did not have access to any of the reasoning models and were, by default, excluded from more advanced models. To learn more about how much better these models are and how you can take advantage of them, keep reading below. As with every model release, the GPT-5 drop was accompanied by benchmark evaluations, in which it earned state-of-the-art scores on across math (AIME 2025), coding (SWE-Bench Verified), and multimodal understanding (MMMU). It even performed competitively on Humanity's Last Exam, a newer benchmark with multi-modal questions in over 100 subjects, such as math, science, and the humanities, as seen below. The company claims it is the strongest coding model yet, being able to create websites, apps, and games from simple text prompts. In particular, OpenAI shares that it has shown improvements in complex front‑end generation and debugging larger repositories. Before the model was released, I watched a live demo of the feature, in which the user created a fully functional web app with interactive elements such as flashcards, a quiz with right and wrong answers, and a game from a simple text prompt. The final product looked sleek, and as someone who has recent experience building webpages, it would have taken me hours to stylize using JavaScript and CSS. GPT-5 appeared to take vibe coding to the next level. Even if you are more of an average GPT-5 user who uses it for writing, you will still reap these benefits. OpenAI claims GPT-5 is the most capable writing collaborator, being able to better tackle tasks that involve "structural ambiguity" such as free verse. Regardless of what you use ChatGPT to write for you, users should see improvements. People have been increasingly reliant on ChatGPT for health-related queries because of its ability to conversationally break down medical jargon, which can often be scary and intimidating. Now the experience is optimized with GPT-5, flagging concerns, asking questions, understanding results, prepping you to ask providers questions, and weighing options. Also: Can AI outdiagnose doctors? Microsoft's tool is 4 times better for complex cases Of course, GPT-5 does not replace a medical professional. OpenAI noted that the model performed the highest on HealthBench, a benchmark evaluation the company itself published earlier this year. External benchmarks for how AI performs in medical scenarios are not yet standardized. One of the biggest improvements available in GPT-5 is that the model is more accurate than any previous reasoning model and has fewer hallucinations, according to OpenAI. The company said GPT-5's responses are 45% less likely to contain a factual error than GPT-4o with web search enabled on anonymized prompts, and 80% less likely to contain a factual error than OpenAI o3. This is a big win, as reasoning models go beyond traditional pattern prediction and are invited to "think," which leaves room for error. OpenAI also added new publicly available benchmarks to test factuality, including LongFact and FActScore, in which GPT-5 with thinking showed a significant drop in hallucinations. GPT-5 (with thinking) also communicates more honestly with the user, sharing when a task is impossible or can't be done. This is important because AI models often offer a plausible-sounding answer instead of admitting they don't know, which can increase the circulation of misinformation. For more on the evaluation results, take a look at the system card. Another brand-new safety feature is called "safe completions," which enables ChatGPT to still answer prompts it would typically refuse. Instead, it will answer, but within the safety boundaries defined by OpenAI, and give a clear explanation of when it can't. Lastly, while not entirely a safety issue, the model is less sycophantic, or effusively agreeable, and uses fewer unnecessary emojis. Also: OpenAI recalls GPT-4o update for being too agreeable GPT-5 is starting to roll out today to all Plus, Pro, Team, and Free users, with access for Enterprise and Edu coming in one week. GPT-5 and GPT-5 mini are available to everyone, including free users and all paid subscribers. However, subscribers still receive tiered perks. For example, included in the $20 per month subscription, ChatGPT Plus subscribers have significantly higher usage limits than free users. Meanwhile, ChatGPT Pro users have unlimited GPT-5 and access to GPT-5 Pro, an even more advanced version of the model included in their $200 per month subscription. The rollout to Enterprise/EDU users will be next week. GPT-5 will be set as the default model for everyday work, replacing all other models for authenticated users. However, paid users will still have the option to select it under the model picker. The release of the model is also helpful for developers, as they can benefit from the increased reliability and accuracy when building. To accommodate this, OpenAI is making GPT-5, GPT-mini, and GPT-5-nano available in the API. Two new parameters, reasoning and verbosity, are also meant to help developers get exactly what they need from their model without overspending. The reasoning parameter makes GPT-5 cheaper for tasks that don't require in-depth thinking, and then the verbosity parameter allows developers to fine-tune just how verbose they want GPT-5 to be. The pricing is cheaper than GPT-4o, and for more information, you can check the blog post.
[10]
With GPT-5, OpenAI Promises Access to PhD-Level' AI Expertise
The new model launches today, even for free users, and OpenAI says it significantly reduces the risk of hallucinations. After a long wait, OpenAI is finally releasing GPT-5, which is designed to offer ChatGPT users access to an AI as smart as a team of PhD experts. "We think you will love using GPT-5 much more than any previous AI," says OpenAI CEO Sam Altman. In a live stream today, OpenAI described GPT-5 as a major upgrade to the intelligence of ChatGPT across a wide range of concentrations, including computer coding, math, language, writing, and even health advice. "With GPT-5, now it's like talking to an expert, a legitimate PhD-level expert in anything, in any area you need," Altman said. The company's tests show that GPT-5 exhibits a much lower risk of hallucinating inaccurate information compared with OpenAI's early models. This includes a lower potential to deceive users by claiming it's completed a task when it actually hasn't. "GPT‑5 not only outperforms previous models on benchmarks and answers questions more quickly, but -- most importantly -- is more useful for real-world queries," the company added in a blog post full of various benchmarks. Our main takeaway from the live stream is that GPT-5 is supposed to be smarter and less error-prone, but we didn't come away wowed. Altman described GPT-5 as an "especially important moment for businesses and developers," suggesting he sees it as a major tool for enterprises. You can test it out yourself. GPT-5 launches today, even for free users on ChatGPT. They'll have access to the GPT-5 model before being downgraded to the GPT-5 mini model once they hit a certain chat limit. "For ChatGPT free-tier users, full reasoning capabilities may take a few days to fully roll out," the company says. "Once free users reach their GPT‑5 usage limits, they will transition to GPT‑5 mini, a smaller, faster, and highly capable model." In addition, free users can expect to receive "hours" of access to ChatGPT's advanced voice mode, rather than only 15 minutes per day. As for paying subscribers, OpenAI says subscribers of ChatGPT Pro -- the $200-per-month plan -- will receive "unlimited access to GPT‑5 and access to GPT‑5 Pro," the smartest and highest performing model. Meanwhile, customers on the $20-per-month Plus plan can expect "significantly" higher access to GPT-5 compared with free users. Paying customers will be able to customize their ChatGPT's personality, which can tweak how the AI responds to questions. "These personalities, available initially for text chat and coming later to Voice, let you set how ChatGPT interacts -- whether concise and professional, thoughtful and supportive, or a bit sarcastic -- without writing custom prompts," the company says. "The four initial options, Cynic, Robot, Listener, and Nerd, are opt-in, adjustable anytime in settings, and designed to match your communication style." Pro, Plus, and Enterprise users will also be able to sync ChatGPT to their Google Calendar, enabling it to learn and coordinate around a customer's schedule. The other important announcement is that GPT‑5 is the "new default in ChatGPT," replacing the company's older models, including GPT‑4o, o3, o4-mini, GPT‑4.1, and GPT‑4.5. GPT-5 itself spans six different models, including four that excel at more complex "thinking," according to OpenAI. But the company says: "In the near future, we plan to integrate these capabilities into a single model." Disclosure: Ziff Davis, PCMag's parent company, filed a lawsuit against OpenAI in April 2025, alleging it infringed Ziff Davis copyrights in training and operating its AI systems.
[11]
OpenAI Launches More Powerful GPT-5 Model for Coding and Writing
OpenAI is rolling out a more powerful and long-awaited new artificial intelligence model called GPT-5, vying to stay ahead of increased competition from rivals in the US and China. GPT-5, set to be unveiled during a livestreamed event on Thursday, is designed to be more capable at coding and creative writing as well as reasoning through complex queries. In a briefing with reporters this week, OpenAI Chief Executive Officer Sam Altman called it a "major upgrade" from the company's prior AI models. For "the first time," he said, "it really feels like talking to an expert in any topic."
[12]
OpenAI launches GPT-5 as the AI industry seeks a return on investment
SAN FRANCISCO, Aug 7 (Reuters) - OpenAI launched on Thursday its GPT-5 artificial intelligence model, the highly anticipated latest installment of a technology that has helped transform global business and culture. OpenAI's GPT models are the AI technology that powers the popular ChatGPT chatbot, and GPT-5 will be available to all 700 million ChatGPT users, OpenAI said. The big question is whether the company that kicked off the generative AI frenzy will be capable of continuing to drive significant technological advancements that attract enterprise-level users to justify the enormous sums of money it is investing to fuel these developments. The release comes at a critical time for the AI industry. The world's biggest AI developers - Alphabet (GOOGL.O), opens new tab, Meta (META.O), opens new tab, Amazon (AMZN.O), opens new tab and Microsoft (MSFT.O), opens new tab, which backs OpenAI - have dramatically increased capital expenditures to pay for AI data centers, nourishing investor hopes for great returns. These four companies expect to spend nearly $400 billion this fiscal year in total. OpenAI is now in early discussions to allow employees to cash out at a $500 billion valuation, a huge step-up from its current $300 billion valuation. Top AI researchers now command $100 million signing bonuses. "So far, business spending on AI has been pretty weak, while consumer spending on AI has been fairly robust because people love to chat with ChatGPT," said economics writer Noah Smith. "But the consumer spending on AI just isn't going to be nearly enough to justify all the money that is being spent on AI data centers." OpenAI is emphasizing GPT-5's enterprise prowess. In addition to software development, the company said GPT-5 excels in writing, health-related queries, and finance. "GPT-5 is really the first time that I think one of our mainline models has felt like you can ask a legitimate expert, a PhD-level expert, anything," OpenAI CEO Sam Altman said at a press briefing. "One of the coolest things it can do is write you good instantaneous software. This idea of software on demand is going to be one of the defining features of the GPT-5 era." One key measure of success is whether the step up from GPT-4 to GPT-5 is on par with the research lab's previous improvements. Two early reviewers told Reuters that while the new model impressed them with its ability to code and solve science and math problems, they believe the leap from the GPT-4 to GPT-5 was not as large as OpenAI's prior improvements. MORE THINKING Nearly three years ago, ChatGPT introduced the world to generative AI, dazzling users with its ability to write humanlike prose and poetry, quickly becoming one of the fastest growing apps ever. In March 2023, OpenAI followed up ChatGPT with the release of GPT-4, a large language model that made huge leaps forward in intelligence. While GPT-3.5, an earlier version, received a bar exam score in the bottom 10%, GPT-4 passed, opens new tab the simulated bar exam in the top 10%. GPT-4's leap was based on more compute power and data, and the company was hoping that "scaling up" in a similar way would consistently lead to improved AI models. But OpenAI ran into issues scaling up. One problem was the data wall the company ran into, and OpenAI's former chief scientist Ilya Sutskever said last year that while processing power was growing, the amount of data was not. He was referring to the fact that large language models are trained on massive datasets that scrape the entire internet, and AI labs have no other options for large troves of human-generated textual data. Apart from the lack of data, another problem was that 'training runs' for large models are more likely to have hardware-induced failures given how complicated the system is, and researchers may not know the eventual performance of the models until the end of the run, which can take months. At the same time, OpenAI discovered another route to smarter AI, called "test-time compute," a way to have the AI model spend more time compute power "thinking" about each question, allowing it to solve challenging tasks such as math or complex operations that demand advanced reasoning and decision-making. GPT-5 acts as a router, meaning if a user asks GPT-5 a particularly hard problem, it will use test-time compute to answer the question. This is the first time the general public will have access to OpenAI's test-time compute technology, something that Altman said is important to the company's mission to build AI that benefits all of humanity. Altman believes the current investment in AI is still inadequate. "We need to build a lot more infrastructure globally to have AI locally available in all these markets," Altman said. Reporting by Anna Tong in San Francisco; Additional reporting by Deborah Sophia in Bengaluru; Editing by Nick Zieminski Our Standards: The Thomson Reuters Trust Principles., opens new tab * Suggested Topics: * Artificial Intelligence Anna Tong Thomson Reuters Anna Tong is a correspondent for Reuters based in San Francisco, where she reports on the technology industry. She joined Reuters in 2023 after working at the San Francisco Standard as a data editor. Tong previously worked at technology startups as a product manager and at Google where she worked in user insights and helped run a call center. Tong graduated from Harvard University.
[13]
GPT-5 is here and it's free for everyone
A couple of days after announcing its first , OpenAI is releasing the long-awaited GPT-5. What's more, you can start using it today, even if you're a free user. With GPT-5, the company is touting across-the-board enhancements, claiming the model is its best yet when it comes to coding, writing, safety, accuracy and more. "GPT-5 is the first time that it really feels like you're talking to an expert in any topic," said OpenAI CEO (and hypeman) Sam Altman during a press briefing the company held before today's announcement. "It reminds me of when the iPhone went from those giant, old pixel [screens] to the Retina Display, and then I went back to using one of those big pixelated things and I was like, 'Wow, I can't believe how bad we had it.'" , Altman said GPT-5 would offer a unified experience for users, and the new model delivers on that promise. For the first time, OpenAI's default offering is a reasoning model, meaning the system is programmed to tackle complex problems by breaking them into smaller parts. Previously, if you wanted to force ChatGPT to use one of OpenAI's reasoning models, you had to select the "Think Longer" option from the prompt bar. This meant most free users didn't even know OpenAI had more capable models. With GPT-5, the company has significantly simplified the ChatGPT experience. On the consumer side of things, there are only three versions of the new model. One of those -- GPT-5 mini -- only crops up when free and Plus users run into their regular GPT-5 usage limit. The other variant, GPT-5 Pro, is, as the name suggests, only available to subscribers of the company's . On the subject of query limits, Plus users can use GPT-5 "significantly" more than those with a free account, while Pro customers can chat with GPT-5 as much as they want. When it comes to reasoning, GPT-5 is much faster than , OpenAI's previous state-of-the-art AI. "It's so fast that I've had the psychological experience of wondering, like, is it really thinking enough? And then it gives a great answer," said Altman. Perhaps more importantly, it suffers from fewer hallucinations, with OpenAI claiming the model delivers more accurate answers than any of its previous reasoning systems. We'll see how GPT-5 responds in real-world use, but if OpenAI has made meaningful improvements here, it would be a big deal; hallucinations have typically been a major weakness of reasoning models, particularly relative to traditional large language counterparts. At the same time, OpenAI says GPT-5 is its safest AI to date. For one, it includes a new feature called Safe Completions. "In the past, we've approached this from a sort of a binary, if we thought that the prompt was safe, we would comply. If we thought it was unsafe, the model would refuse," said Alex Beutel, safety research lead at OpenAI. "This worked well, but as a challenge that there can be kind of carefully worded prompts that could be confusing. So if someone says how much energy is needed to ignite some specific material that could be an adversary trying to get around the safety protections and cause harm, or it could be a student asking a science question to understand the physics of this material." With Safe Completions, GPT-5 will try to give the most helpful answer within the safety constraints OpenAI has imposed on it. In tricky situations like the one Beutel outlined above, the model will only provide high-level information that can't be used to harm anyone. "On average, the system is both safer and more helpful for users, and we think that'll be much better," Beutel added. Additionally, when it comes to health-related questions, GPT-5 is better at flagging concerns and suggesting questions the user should ask of their healthcare provider. It will also answer those prompts more precisely, thanks to the ability to adapt to the person's knowledge level and geography. On top of everything else, OpenAI says GPT-5 is its best model for coding yet. It's supposedly a better writer too, with the company promising the chatbot is better at translating your drafts into "compelling, resonant" copy. Alongside GPT-5, OpenAI is adding a handful of new features to ChatGPT. To start, users can now choose a color for their chats, with a few exclusive options available for paying customers. OpenAI has also made it easier to connect ChatGPT to Gmail, Google Calendar and Google Contacts. Once you enable the connections, the chatbot will know when to automatically reference your Google accounts; you won't need to select anything before you start chatting. OpenAI will begin rolling out this feature to Pro subscribers starting next week, with availability for other users to follow. Over in the Custom Instructions pane, where you can write system prompts to tweak how ChatGPT interacts with you, OpenAI is introducing a handful of pre-set personalities. The four options -- cynic, robot, listener and nerd -- are available as part of a research preview, and can be changed or disabled at any time. Last but not least, OpenAI is releasing an updated version of its Advanced Voice feature the company . OpenAI says the tool is better at understanding instructions and adapting its speaking style to the moment. As part of this change, OpenAI is retiring Standard Voice Mode. In practice, that means the company can now offer a better voice experience to everyone since it doesn't need to fall back on Standard Voice Mode, which isn't natively multi-modal like Advanced Voice and therefore worse at understanding the nuances of human speech. If you're wondering where this leaves OpenAI on the path toward artificial general intelligence, Altman had this to say when asked about the topic. "I kind of hate the term AGI, because everyone at this point uses it to mean a slightly different thing, but [GPT-5] is a significant step forward towards models that are really capable. We're still missing something quite important," he said, noting GPT-5 can't continuously learn on its own. "But the level of intelligence here, the level of capability, it feels like a huge improvement. Certainly, if I could go back five years before GPT-3 and you told me we have this now, I'd be like that's a significant fraction of the way to something very AGI-like."
[14]
OpenAI launches new GPT-5 model for all ChatGPT users
OpenAI on Thursday announced GPT-5, its latest and most advanced large-scale artificial intelligence model. The company is making GTP-5 available to everyone, including its free users. OpenAI said the model is smarter, faster and "a lot more useful," particularly across domains like writing, coding and health care. "I tried going back to GPT-4, and it was quite miserable," OpenAI CEO Sam Altman said in a briefing with reporters. Since launching its AI chatbot ChatGPT in 2022, OpenAI has rocketed into the mainstream. The company said it expects to hit 700 million weekly active users on ChatGPT this week, and it is in talks with investors about a potential stock sale at a valuation of roughly $500 billion, as CNBC previously reported. OpenAI said GPT-5's hallucination rate is lower, which means the model fabricates answers less frequently. The company said it also carried out extensive safety evaluations while developing GPT-5, including 5,000 hours of testing. Instead of outright refusing to answer users' questions if they are potentially risky, GPT-5 will use "safe completions," OpenAI said. This means the model will give high-level responses within safety constraints that can't be used to cause harm. "GPT-5 has been trained to recognize when a task can't be finished, avoid speculation and can explain limitations more clearly, which reduces unsupported claims compared to prior models," said Michelle Pokrass, a post-training lead at OpenAI.
[15]
OpenAI releases GPT-5, a potential barometer for whether artificial intelligence hype is justified
SAN FRANCISCO (AP) -- OpenAI has released the fifth generation of the artificial intelligence technology that powers ChatGPT, a product update that's being closely watched as a measure of whether generative AI is advancing rapidly or hitting a plateau. GPT-5 arrives more than two years after the March 2023 release of GPT-4, bookending a period of intense commercial investment, hype and worry over AI's capabilities. In anticipation, rival Anthropic released the latest version of its own chatbot, Claude, earlier in the week. Expectations are high for the newest version of OpenAI's flagship model because the San Francisco company has long positioned its technical advancements as a path toward artificial general intelligence, or AGI, a technology that is supposed to surpass humans at economically valuable work. It is also trying to raise huge amounts of money to get there, in part to pay for the costly computer chips. OpenAI started in 2015 as a nonprofit research laboratory to safely build AGI and has since incorporated a for-profit company with a valuation that has grown to $300 billion. The company has tried to change its structure since the nonprofit board ousted its CEO Sam Altman in Nov. 2023. He was reinstated days later and continues to lead OpenAI. It has run into hurdles escaping its nonprofit roots, including scrutiny from the attorneys general in California and Delaware, who have oversight of nonprofits, and a lawsuit by Elon Musk, an early donor to and founder of OpenAI. Most recently, OpenAI has said it will turn its for-profit company into a public benefit corporation, which must balance the interests of shareholders and its mission.
[16]
OpenAI claims new GPT-5 model boosts ChatGPT to 'PhD level'
OpenAI has highlighted GPT-5's ability to create software in its entirety and demonstrate better reasoning capabilities - with answers that show workings, logic and inference. The company claims it has been trained to be more honest, provide users with more accurate responses and says that, overall, it feels more human. According to Altman, the model is "significantly better" than its predecessors. "GPT-3 sort of felt to me like talking to a high school student... 4 felt like you're kind of talking to a college student," he said in a briefing ahead of Thursday's launch. "GPT-5 is the first time that it really feels like talking to an expert in any topic, like a PhD-level expert." For Prof Carissa Véliz of the Institute for Ethics in AI, however, GPT-5's launch may not be as significant as its marketing may suggest. "These systems, as impressive as they are, haven't been able to be really profitable," she said, also noting that they can only mimic - rather than truly emulate - human reasoning abilities. "There is a fear that we need to keep up the hype, or else the bubble might burst, and so it might be that it's mostly marketing." The BBC's AI Correspondent Marc Cieslak gained exclusive access to GPT-5 before it's official launch. "Apart from minor cosmetic differences the experience was similar to using the older chatbot: give it tasks or ask it questions by typing a text prompt. It's now powered by what's called a reasoning model which essentially means it thinks harder about solving problems, but this seems more like an evolution than revolution for the tech." The company will roll out the model to all users from Thursday. In the coming days it will become a lot clearer whether it really is as good as Sam Altman claims it is.
[17]
OpenAI Aims to Stay Ahead of Rivals With New GPT-5 Technology
The A.I. start-up said its new flagship technology was faster, more accurate and less likely to make stuff up. ChatGPT is getting another upgrade. On Thursday, OpenAI unveiled a new flagship A.I. model, GPT-5, and began sharing the technology with the hundreds of millions of people who use ChatGPT, the company's online chatbot. During a briefing with journalists, OpenAI executives called GPT-5 a "major upgrade" over the systems that previously powered ChatGPT, saying the new technology was faster, more accurate and less likely to "hallucinate," or make stuff up. "It feels significantly better in obvious ways and in subtle ways," OpenAI's chief executive, Sam Altman, said. "GPT-5 is the first time that it feels like talking to an expert in any topic -- a Ph.D.-level expert." Since launching the A.I. boom in late 2022 with the release of ChatGPT, OpenAI has consistently improved the technology that underpins its chatbot. This began with the release of the company's GPT-4 technology in the spring of 2023 and continued through a series of A.I. models that could listen, look and talk and approximate the way people reason through complex problems. OpenAI's many rivals, including Google, Meta, the start-up Anthropic and China's DeepSeek, have released similar technologies. This is the first time that OpenAI has used a so-called reasoning model to power the free version of ChatGPT. Unlike the previous technologies, a reasoning model can spend time "thinking" through complex problems before settling on an answer. "For most people on ChatGPT, this is their first introduction to reasoning," said Nick Turley, the OpenAI vice president who oversees ChatGPT. "It just knows when to 'think.'" (The New York Times has sued OpenAI and its partner, Microsoft, accusing them of copyright infringement of news content related to A.I. systems. OpenAI and Microsoft have denied those claims.) OpenAI said that the technology "feels more human" than previous models and that it allowed even novices to build simple software apps from short text prompts. One OpenAI engineer asked the system to generate an online app that could help people learn French, and it created this app in minutes. Mr. Altman called the system a "significant step" along the path to the ultimate goal of the company and its rivals: artificial general intelligence, or A.G.I., a machine that can do anything the human brain can do. But he also acknowledged that it lacked many of the key ingredients needed to build such a machine. Many experts say there is no clear path to developing A.G.I. Earlier this week, OpenAI said it was "open sourcing" two other A.I. models that can power online chatbots, freely sharing the technology with researchers and business across the globe. Since unveiling ChatGPT three years ago, the company has mostly kept its technology under wraps. If people use these open-source models, OpenAI hopes they will also pay for its more powerful products. In addition to offering a free chatbot via the internet, OpenAI sells access to a more powerful chatbot for $20 a month and sells a wide range of A.I. technologies to businesses and independent software developers. The company is not yet profitable. It plans to raise $40 billion this year and is on a pace to pull in revenues of $20 billion by year's end.
[18]
OpenAI made a big upgrade to ChatGPT. Here's what to know about GPT-5.
The company behind ChatGPT says the chatbot is set to get better at writing and answering health questions. SAN FRANCISCO -- ChatGPT maker OpenAI on Thursday released GPT-5, an upgraded version of the artificial intelligence technology that powers its popular chatbot, promising better results on tasks including answering health questions, writing and generating computer code. ChatGPT has more than 700 million users every week, OpenAI has said, and is used by people to help out at home, work and school. The company has often claimed its technology is advancing quickly toward the point it will be able to match or exceed humans at every possible task. Executives said in a news briefing that GPT-5 does not perform at that level. How much does it cost to access GPT-5? All ChatGPT users will get access to GPT-5, even those using the free version. But only those with a $200 a month "Pro" subscription get unlimited access to the newly released system. GPT-5 will be the default mode on all versions. Users not paying for ChatGPT will only be able to ask a certain number of questions answered by GPT-5 before the chatbot switches back to using an older version of OpenAI's technology. How will GPT-5 change ChatGPT? GPT-5 is faster to respond to questions than OpenAI's previous offerings and less likely to "hallucinate" or make up false answers, OpenAI executives said at a news briefing ahead of its release. It gives ChatGPT "better taste" when generating writing, said Nick Turley, who leads work on the chatbot. OpenAI's new AI software can also answer queries using a process dubbed "reasoning" that shows the user a series of messages attempting to break down a question into steps before giving its final answer. "GPT-5 is the first time that it really feels like talking to an expert, a PhD-level expert," OpenAI CEO Sam Altman said of the experience of using the upgraded ChatGPT. Altman said GPT-5 is particularly good at generating computer programming code, a feature that has become a major selling point for OpenAI and rival AI developers and transformed the work of programmers. In a demo, the company showed how two paragraphs of instruction was enough to have GPT-5 create a simple website offering tutoring in French, complete with a word game and daily vocabulary tests. Altman predicted that people without any computer science training will one day be able to quickly and easily generate any kind of software they need to help them at work or with other tasks. "This idea of software on demand will be a defining era of the new GPT-5 era," Altman said. <b>What does it mean for an AI chatbot to "reason?"</b> GPT-5 could give many people their first encounter with AI systems that attempt to work through a user's request step-by-step before giving a final answer. That so-called "reasoning" process has become popular with AI companies because it can result in better answers on complex questions, particularly on math and coding tasks. Watching a chatbot generate a series of messages that read like an internal monologue can be alluring, but AI experts warn users not to confuse the technique with a peek into AI's black box. The self-chatter doesn't necessarily reflect an internal process like that of a human working on a problem, but designing chatbots to create what are sometimes dubbed "chains of thought" forces the software to allocate more time and energy to a query. OpenAI released its first reasoning model in September for its paying users but Chinese start-up DeepSeek in January released a free chatbot that made its "chain of thought" visible to users, shocking Silicon Valley and temporarily tanking American tech stocks. The company said ChatGPT will now automatically send some queries to the "reasoning" version of GPT-5, depending on the type of conversation and complexity of the questions asked. Is GPT-5 the "super intelligence" or "artificial general intelligence" OpenAI has promised? No. Tech leaders have for years been making claims that AI is improving so fast it will soon become able to learn and perform all tasks that humans can at or better than our own ability. But GPT-5 does not perform at that level. Super intelligence and artificial general intelligence, or AGI, remain ill-defined concepts because human intelligence is very different from the capabilities of computers, making comparisons tricky. OpenAI CEO Altman has been one of the biggest proponents of the idea that AI capabilities are increasing so rapidly they will soon revolutionize many aspects of society. "This is a significant step forward," Altman said of GPT-5. "I would say it's a significant fraction of the way to something very AGI-like."
[19]
ChatGPT-5 is here -- OpenAI's most powerful AI model just changed everything
OpenAI has officially launched GPT-5, its most advanced AI model to date. In a pre-briefing, Sam Altman shared that once he tried GPT-5 he had a hard time going back to GPT-4, likening the revert to using an old iPhone. Designed to be more useful, more accurate and more human than any version before it, this is by far OpenAI's smartest and most efficient model. GPT-5 is now rolling out in ChatGPT and the API, with the company describing it as a "unified system" built to deliver the best possible answer every time; no model switching required. This marks a major shift from the current GPT-4/4o experience and introduces a more seamless way to interact with AI. Smarter, safer and more proactive OpenAI says GPT-5 brings noticeable improvements in reasoning, accuracy and reliability. It hallucinates less, adapts better to your context and introduces a new feature called "safe completion." In other words, the model won't blindly comply with risky prompts or flatly refuse either. Instead, it aims to give the most helpful response within safety boundaries, offering clear reasoning when it can't help. A major leap in coding and workflows GPT-5 is OpenAI's strongest coding model ever. It can create entire websites, apps, and games, even for users who don't know how to code. Developers can expect smarter debugging, cleaner frontend design and more agentic behavior, with tools like free-form function calling, verbosity control and reasoning effort toggles. For non-technical users and casual users, GPT-5 also enhances writing workflows, supports health-related questions with improved contextual understanding and is more proactive in helping users weigh options or understand complex information. OpenAI notes that it doesn't replace medical professionals, but it's now better equipped to support health-related discussions. New customization and voice upgrades ChatGPT now lets users customize the look and tone of their experience. You can change chat colors and choose from new preset personalities like "Cynic," "Listener" or "Robot." Voice interactions have also improved with adaptive tone control and higher usage limits for paid users. OpenAI is unifying all users under this new voice mode and retiring the previous "Standard Voice" within 30 days. Connect Gmail, Google Calendar and more Another standout feature is Google integration. Users will soon be able to connect Gmail, Calendar and Contacts directly to ChatGPT. Once enabled, the assistant will intelligently know when to reference your data, without having to be explicitly told. Available now and free for all users GPT-5 is rolling out starting today to all ChatGPT users and is available for everyone for free. Free users will get access to GPT-5 and GPT-5 mini, though usage is capped. Plus and Pro users will see significantly higher usage limits, with unlimited access available to Pro subscribers. GPT-5 will also be the default model for enterprise and EDU accounts next week. On the developer side, GPT-5 is launching with three model variants, gpt-5, gpt-5-mini, and gpt-5-nano, with improved latency and cost options. Pricing is more affordable than GPT-4o, with caching discounts built in. Final thoughts With GPT-5, OpenAI is redefining what AI can be in everyday life. Whether you're building software, writing reports, managing a schedule, or just trying to get a clearer answer to a complex question, GPT-5 promises to take it all in one with ease and accuracy. Follow Tom's Guide on Google News to get our up-to-date news, how-tos, and reviews in your feeds. Make sure to click the Follow button.
[20]
GPT-5 free tier - 3 cool features you can try now for nothing
At today's ChatGPT-5 launch event, we got to see exactly why OpenAI is so excited about the new version of ChatGPT. It's faster, more intuitive, less likely to hallucinate, and can integrate with your Gmail in even more useful ways to know more about you. In fact, at one point, the OpenAI presenters actually got ChatGPT-5 to write a eulogy to the previous ChatGPT-4o model as a demonstration about how much better its writing capabilities had become. The eulogy was warranted because OpenAI isn't restricting access to ChatGPT-5 to Plus and Pro subscribers - it is giving it out to free tier users too, so everybody now gets access to a better ChatGPT. That effectively means the end for ChatGPT-4o. The vast majority of today's presentation was about how much better ChatGPT-5 was at writing code than previous versions, and also how much better it was at enterprise applications, like law, health care, education, and finance. Of course, most of us simply use ChatGPT to help us get things done in our daily lives, but there were plenty of improvements that are worth noting for everyday users, too, and they're mostly available in the free tiers. Let's take a look at what you get access to without paying. As with GPT‑4o, the difference between free and paid access to GPT‑5 is usage volume. At some point on the free tier, your access to GPT-5 will reach the usage limit, and then you'll be kicked back to GPT-5 mini. Plus subscribers will get significantly higher usage volume before that happens. That means you'll get access to the improved writing ability in GPT-5. You'll find that its prose has a lot more rhythm and beat to it than the prose in GPT-4o did. With the free tier, you still only get limited access to ChatGPT-5's voice mode (along with file uploads, image creation, and data analysis), but you should get more access per day now than you did before. What used to be called 'Advanced voice mode' is now called 'ChatGPT voice'. OpenAI says that "Standard Voice Mode retires on September 9, 2025, unifying all users on ChatGPT Voice". ChatGPT Voice is better than before, with more natural-sounding conversations. Hit the voice button on the mobile app for ChatGPT to give it a go right now. You've always been able to change the theme of ChatGPT from light to dark to System, but now you can add accent colors to your chats in the Settings. These apply to elements in ChatGPT-5 like conversation bubbles and highlighted text. This is the first time color has been used in the ChatGPT interface, and it takes some getting used to, but I think this is a feature you'll come to enjoy. ChatGPT-5 has better memory, reduces hallucinations, and as a free user, you'll get all that too. So what don't you get? Well, as before, access to the Sora AI video generator is restricted to Plus and Pro users, as is ChatGPT Agent. And in terms of new models, the super-powerful ChatGPT-5 Pro is only available to Pro users. If you're frantically hitting the refresh button on your ChatGPT page waiting for access to GPT-5 then you're not alone right now, but it is on the way. A message on the OpenAI website says: "We are gradually rolling out GPT-5 to ensure stability during launch. Some users may not yet see GPT-5 in their account as we increase availability in stages."
[21]
GPT-5 accidentally leaked hours before OpenAI reveals the future of ChatGPT
The next generation of ChatGPT, titled GPT-5, is expected to be revealed later today in an OpenAI livestream - but a leak on GitHub appears to have revealed everything the AI pioneers will unveil during the event. The leaked information appeared on GitHub, highlighting the different iterations of GPT-5, which is describe as "OpenAI's most advanced model, offering major improvements in reasoning, code quality, and user experience." The GitHub blog leak has since been taken down, but can be easily accessed via the Internet archive. The post highlights the following four new models: We expect to get more information on these new models during today's livestream, which begins at 10am PT / 1pm ET / 6pm BST. While none of the models in the leak seem to hint at the model selection-less future we were hoping for, I'm still optimistic that GPT-5 ushers in a future where OpenAI's complex naming scheme isn't a point of friction for end users. Sam Altman has promised in the past that ChatGPT's future would incorporate all models to determine which one can answer the user's prompt most efficiently. While this leak doesn't appear to showcase this future, Altman might reveal this innovation later today. Stay tuned to TechRadar for all the GPT-5 news as it's announced. We'll be live-blogging throughout the day, documenting any information we find on the future of ChatGPT in the build-up to the livestream.
[22]
How to watch OpenAI's livestream and expected GPT-5 announcement today
OpenAI announced a livestream for 10 a.m. PT (1 p.m. ET) on Aug. 7, 2025, writing on X: "LIVE5TREAM THURSDAY 10AM PT." You might have noticed a typo -- the "S" usually found in the word "livestream" is actually a 5. That's right -- OpenAI appears to be teasing the highly anticipated launch of its new AI model GPT-5, including rumored features like better coding abilities, larger context windows, improved video generation with Sora, improved memory, and more. This Tweet is currently unavailable. It might be loading or has been removed. Despite recent reports that GPT-5 was being delayed yet again, all signs point to "Go" for an August release, and you can see for yourself during the livestream later today. We've seen a steady stream of GPT-leaks this week, with Microsoft seemingly confirming the imminent arrival of new models such as GPT-5, GPT-5-mini, and GPT-5-nano, per Bleeping Computer. To watch the livestream, head to OpenAI's YouTube channel, where the tech giant typically hosts its streams. As of this writing, a livestream video isn't online just yet, but it will likely show up closer to launch time.
[23]
OpenAI's GPT-5 is here and free for all ChatGPT users
It's finally here. Credit: Ian Moore / Mashable Composite; OpenAI On Thursday, OpenAI launched what it describes as its smartest and fastest model yet during a livestream event. In a press briefing, OpenAI CEO Sam Altman detailed the foundation model's capabilities. "GPT-3 sort of felt to me like talking to a high school student... GPT-4 felt like you're kind of talking to a college student, GPT-5 is the first time that it really feels like talking to an expert in any topic," said Altman, who compared it to a "PhD level expert." OpenAI says GPT-5 has improved expertise across coding, math, science, writing, and healthcare. Crucially, the company also claims the new large language model has a lower hallucination rate than o3, which was an unsettling 33 percent. But the key takeaway, Altman said, is overall improvements in both obvious and subtle ways. "I tried going back to GPT-4, and it was quite miserable," he said. GPT-5 will roll out today to ChatGPT free, Pro, and Plus users, but is not yet available as of this writing. OpenAI says free users will have the strictest rate limits, Pro users will get higher rate limits, and Plus users will have unlimited use. Let the testing begin. GPT-5 also delivers a new user experience, combining multiple tools into one and expanding access to OpenAI's more advanced AI tools. Users fatigued by the confusing model picker will no longer need to manually select the right model for a task. GPT-5 does away with that and combines reasoning capabilities from previous models into one unified experience. Head of ChatGPT Nick Turley underscored that this is the first time free users will have access to reasoning, and "they don't have to think about it, because GPT-5 just knows when to think." "The vibes of this model are really good, and I think that people are really going to feel that," said Turley, "especially average people who haven't been spending their time thinking about models." Speaking of vibes, OpenAI demoed a vibe-coding scenario where GPT-5 created a language-learning app with quizzes and games, all from a short natural language prompt. "[GPT-5] really puts the power of beautiful and effective coding in everyone's hands," said Yann Dubois, a GPT-5 researcher. GPT-5 arrives after years of anticipation and rumored setbacks. It's been over two years since OpenAI launched GPT-4, the large language model that gave the already-popular ChatGPT a turbo boost in capabilities and fueled commercial adoption of AI models. But the GPT-5 launch has been repeatedly delayed amidst reports that the OpenAI team was struggling to achieve performance gains significant enough to warrant an entirely new model. ChatGPT is currently the most well-known AI chatbot, with 700 million weekly active users, according to OpenAI. This means OpenAI has an outsized impact on the industry as a whole. Mashable is just beginning to test GPT-5, and we'll have more to share on its features and capabilities in the days and weeks ahead. For now, it's too early to say if GPT-5 will silence AI critics, who say agentic AI tools have failed to meet expectations. Some AI experts also believe that scaling laws -- improving AI models through sheer data and computing power -- is hitting a wall, suggesting the current path to AGI is longer than what OpenAI claims, if it's even possible. But Altman said scaling laws "absolutely still hold" in the briefing. "We keep finding new dimensions to scale on, we can still scale the pre-training beautifully and get huge returns," he said. "Obviously, we have to invest in compute at like an-eye watering rate to get that, but we intend to keep doing it." Meanwhile, a troubling new issue has emerged where users have experienced what's described as "ChatGPT psychosis" and worsening existing mental health issues. A more advanced model raises questions about how users will relate to it and OpenAI's responsibility to mitigate harmful interactions. "We take these issues extremely seriously," said Turley, who highlighted new features announced this week for healthier ChatGPT use. Turley said OpenAI has been collaborating with mental health experts, making improvements on issues like emotional reliance and ChatGPT sycophancy, and will have "a lot more to share soon."
[24]
OpenAI says latest ChatGPT upgrade is big step forward but still can't do humans' jobs
Though GPT-5 model has better coding and writing abilities it is not yet able to 'continuously learn' OpenAI has claimed to have taken a "significant step" towards artificial general intelligence (AGI) with the launch of its latest upgrade to ChatGPT, but has admitted there are still "many things" missing in its quest to create a system able to do humans' jobs. The startup said its GPT-5 model, the underlying technology that will power its breakthrough AI chatbot, represents a big upgrade on its predecessors in areas such as coding and creative writing - and is also a lot less sycophantic. It said the upgrade was being made available to all of ChatGPT's 800 million users immediately. Sam Altman, OpenAI's chief executive, called the model a "significant step forward" to achieving the theoretical state of AGI, which the startup defines as a highly autonomous system that outperforms humans at most economically valuable work - or in other words, can do their jobs. However, Altman admitted GPT-5 had not reached that goal yet. "[It is] missing something quite important, many things quite important," said Altman, pointing to the model's inability to "continuously learn". Altman said GPT-5 was "generally intelligent" and a "significant step on the path to AGI" but had not reached it yet by most people's definition. "I think the way that most of us define AGI, we're still missing something quite important, many things quite important. But one big one is ... this is not a model that continuously learns as it's deployed from things it finds, which is something that, to me, feels like it should be part of AGI," he said, adding that, nonetheless, it was a "huge improvement" on its predecessors. AGI's theoretical capabilities, and the determination of heavily backed tech companies to achieve it, has led to predictions from AI executives that white-collar jobs - from lawyers to accountants, doctors and bankers - will be wiped out by looming advances in the technology. Dario Amodei, the boss of AI developer Anthropic, has warned the technology could replace half of all entry-level office jobs in the next five years. OpenAI said the main improvements in GPT-5 included: fewer factual errors, or hallucinations; better software coding, allowing it to create functional websites and apps; increased capability at creative writing; and, rather than "refusing" a prompt that breaches its guidelines outright, the model will instead try to give the most helpful response possible within safety guidelines, or at least explain why it cannot help. The agent feature in ChatGPT - which carries out tasks such as finding restaurant availability and shopping online - will also be able to access users' Gmail, Google calendar and contacts, if given permission. As with its predecessors, GPT-5 can generate voice, image and text and can deal with queries in those formats too. OpenAI said the upgraded ChatGPT would be better at answering health-related questions and would be more proactive at "flagging potential concerns" - such as serious physical or mental illness. The startup stressed the chatbot was not a replacement for professional help, amid concerns AI tools could worsen the condition of people vulnerable to psychosis. Nick Turley, the head of ChatGPT at OpenAI, said the model showed "significant improvements on sycophancy", after the startup's admission this year that the chatbot's most sophisticated model had become too agreeable and could distress users or make them uncomfortable. The latest model release comes as tech firms pour multibillion-dollar sums into their attempts to achieve AGI. On Tuesday, Google's AI unit outlined its latest step towards AGI by showcasing an unreleased "world model", while last week Mark Zuckerberg, the CEO of Facebook parent Meta, said the development of superintelligence - another theoretical state of AI even more powerful than AGI - was "now in sight". Investors' faith in the potential for further breakthroughs, and AI's ability to transform modern economies, has also spurred a valuation boom in companies such as San Francisco-based OpenAI. It was reported on Wednesday that OpenAI is in early talks about a sale of shares held by current and former employees that would value it at $500bn, overtaking Elon Musk's SpaceX. Although OpenAI also released two open models this week and has a free version of ChatGPT, the company makes its revenues from charging for subscriptions to the most powerful versions of the chatbot and from integrating its models into businesses' IT systems. Use of the free version of GPT-5-backed ChatGPT will be capped, while users on the $200-a-month Pro package will get unlimited access.
[25]
The 7 best new GPT-5 features to try right away
ChatGPT is about to get a whole lot better. OpenAI has finally taken the wraps off of GPT-5, its latest and greatest large language model. GPT-5 has been rumored and speculated about for as long as GPT-4 has been around, and for good reason. The new foundational OpenAI model packs a ton of improvements that will make ChatGPT more accurate, more customizable, and more helpful. For starters, GPT-5 is smarter, faster, and more accurate -- which, yes, means fewer "hallucinations," which occur when ChatGPT makes up information. As one OpenAI leader described it during the announcement livestream, it's like having a subject matter expert on anything in your pocket. GPT-5 is also being positioned as a model that can do everything. It can write, code, create interactive apps, check your email, and conduct deep research. Curious about what's on offer in GPT-5 and whether or not it will actually improve the ChatGPT experience? Like the rest of the world, we're still investigating GPT-5 for ourselves. In the meantime, here are all the best new features to try with GPT-5, which rolls out today for free to all ChatGPT users. GPT-5 integrates all of the various models on offer by ChatGPT. That means instead of picking the right model for the task at hand, ChatGPT will pick the best model(s) for your prompt. Simply ask GPT-5, and it will decide how to approach a problem on its own. OpenAI is introducing additional models along with GPT-5, which is the do-it-all model that will be OpenAI's flagship model going forward. There's also GPT-5 Mini and GPT-5 Nano, which are more cost-effective than GPT-5, and will be accessible to developers who want to pay less. Users will automatically transition to using these smaller models when they hit their GPT-5 usage limits; however, OpenAI says they're still very capable models -- and even more capable than GPT-o3 in many cases. OpenAI is positioning GPT-5 as being your own personal coder, and the company says GPT-5 is significantly better at writing code. If you're a programmer, then OpenAI wants GPT-5 to be your new coding partner. But what about the rest of us? The so-called vibe coders of the world? During a live demo, company leaders showed how to use GPT-5 to create custom, interactive applications on demand using a simple prompt. So, you could think of an app or a piece of software that you wish you had, and simply ask ChatGPT to make it. Of course, in the real world GPT-5 probably won't be quite as seamless as that -- it's still likely to make errors, and won't get everything right the first time. But if you're interested in so-called "vibe-coding," GPT-5 looks like a big leap forward. Because GPT-5 integrates all of OpenAI's models into one AI chatbot -- and because it's free to all ChatGPT users -- free users can finally gain access to OpenAI's most advanced reasoning models. Reasoning models can conduct multi-part searches based on your prompt, then synthesize vast amounts of data into a response. And according to OpenAI CEO Sam Altman, GPT-5 can do this at a PhD level. We haven't tested this out for ourselves yet, and we're always nervous about hallucinations. However, if you only use the free ChatGPT tier, then this kind of deep research is definitely worth trying. But GPT-5 doesn't just do research. During the livestream, OpenAI leaders showed how GPT-5 could create custom visualizations, interactive apps, games, quizzes, and other tools to help you learn about the subject at hand. OpenAI is also debuting four new preset personalities in ChatGPT. These can inform how ChatGPT will interact with you as you chat and are set on a chat-by-chat basis. Initially, there will be four options: You don't actually have to use any of these personalities if you don't want to, but you can use them for specific use cases as needed. You can also adjust them in ChatGPT settings. OpenAI says the new personalities were specifically designed to reduce the sycophancy issue with GPT-4o. Voice Mode makes it easy to get information from ChatGPT without having to type every single word. With GPT-5, Voice Mode is getting better, OpenAI says. The company says Voice Mode will now adapt and better understand your instructions so that it can tailor how it speaks depending on the moment. All users will have access to this improved ChatGPT Voice Mode, including free users, though of course paid users will have higher usage limits. OpenAI is also making Voice Mode more accessible, and it will now be available in custom GPTs. The lack of advanced voice mode in custom GPTs has been sorely missing, so this update will be super helpful for those who use custom GPTs for different tasks. The ability to connect ChatGPT to external apps like Google Drive has been helpful for those who want their AI assistant to be able to reference information in their Google account. Now, those connections are getting much more helpful. You'll be able to enable Gmail, Google Calendar, and Google Contacts connections with ChatGPT, and going forward, ChatGPT will automatically pull information from them and reference them as necessary. That means that you won't have to explicitly select the connections in a chat each time. For some, of course, this might raise privacy concerns, but rest assured, you don't have to connect these apps or services if you don't want to. At launch, you might not be able to connect them anyway. Initially, this feature will only be available to Pro users and will start rolling out next week. Other tiers will get it too, but we don't know exactly when. Apart from simply being underpinned by a better model, GPT-5 also heralds a few new features for ChatGPT that can even further enhance the experience of using it. An example of that is the fact that you can now choose colors for different chats. That may sound like a small update (and to be clear, it is), but this feature could help you much better organize your chats and more easily jump between them. OpenAI says GPT-5 can write better, code better, and even better inform users on issues around health. (To be clear, you should still consult with a medical professional before you make any health-related decisions. ChatGPT does not replace your doctor.) But there are some ways in which GPT-5 is not better than existing OpenAI large language models. For example, it has a wider context window than GPT-4o at 256,000 tokens, but that's still nowhere near as impressive as GPT-4.1, which has a one-million-token context window. Thankfully, OpenAI says it will continue to support GPT-4.1, which might be a better model for those times in which you need ChatGPT to process massive amounts of information at a time.
[26]
OpenAI is bringing 4 new 'personalities' to ChatGPT to reduce sycophancy
As part of today's GPT-5 announcement, OpenAI also revealed that it's bringing four new personality options to ChatGPT. On Thursday, OpenAI revealed its newest foundational large-language model, GPT-5. The new model from OpenAI integrates many ChatGPT features and reasoning models into a single system, making it easier for users to conduct deep research, create custom applications, and complete tasks. In a blog post, OpenAI said users will also have access to four personality options for the AI chatbot: Cynic, Robot, Listener, and Nerd. "We're also launching a research preview of four new preset personalities for all ChatGPT users, made possible by the improvements on steerability. These personalities, available initially for text chat and coming later to Voice, let you set how ChatGPT interacts -- whether concise and professional, thoughtful and supportive, or a bit sarcastic -- without writing custom prompts." OpenAI says these new personalities are opt-in. They're also adjustable, letting users fine-tune the chatbot's personality to their needs. Critically, OpenAI says these new ChatGPT personalities will help address the sycophancy problem that plagued recent releases from the AI company. The most recent iterations of ChatGPT developed a deserved reputation as a suck up. That proved to be very problematic, with some users reporting that ChatGPT encouraged delusions and worsened mental health crises, as the New York Times reported. Combined with the high hallucination rates of some ChatGPT models, it was a recipe for a poor user experience. "All of these new personalities meet or exceed our bar on internal evals for reducing sycophancy," reads the blog post. "Overall, GPT‑5 is less effusively agreeable, uses fewer unnecessary emojis, and is more subtle and thoughtful in follow‑ups compared to GPT‑4o," the post states. "It should feel less like 'talking to AI' and more like chatting with a helpful friend with PhD‑level intelligence." As of this writing, GPT-5 is not yet available, though it will be rolling out today.
[27]
ChatGPT jumps a level with OpenAI's major GPT-5 update
Why it matters: The Thursday release of GPT-5 brings together traditional and "reasoning" models and ups the ante in the race toward so-called artificial general intelligence (AGI). Driving the news: OpenAI is making the new model available Thursday for free and paid users, with enterprise and educational customers getting access starting next week. * GPT-5 will have fewer hallucinations and be better at coding and health-related tasks, among other improvements, the company said. Yes, but: Free ChatGPT users will have a limited capacity for GPT-5 use. * Once they hit their limit, they'll be switched to a smaller GPT-5 mini model. * Paying subscribers will be able to use GPT-5 as their default model, while Pro users will have unlimited GPT-5 queries and access to GPT-Pro, which uses additional computing capacity. Zoom in: OpenAI said GPT-5 sets new highs on a range of benchmarks and cited praise from a variety of companies that had early access, including Windsurf, Vercel, Notion and Manus. * Altman told a press briefing that the new model not only excels at coding but at taking software projects from idea to usable code in a single step. * "This idea of software on demand will be a defining part of the new GPT-5 era," he said. Zoom out: Altman said that GPT-5 shows signs of broad intelligence but lacks the ability to learn and improve on its own -- a capacity that, he added, is likely a key component of AGI. * After you use GPT-5, Altman said, going back to an earlier model is like reverting to a lower-resolution display after you've grown used to an iPhone's Retina display. Between the lines: Users will also be able to customize GPT-5's personality as part of a feature in "research preview." * The four initial options are "Cynic," "Robot," "Listener" and "Nerd." The new model displays less sycophantic behavior than its predecessors, OpenAI said in a blog post. * "Overall, GPT‑5 is less effusively agreeable, uses fewer unnecessary emojis, and is more subtle and thoughtful in follow‑ups compared to GPT-4o. It should feel less like 'talking to AI' and more like chatting with a helpful friend with PhD‑level intelligence." Threat level: As it did with its recently released ChatGPT Agent, OpenAI is proceeding as if GPT-5 ranks "high" on its risk scale for biological threat capability and has added additional safeguards. What we're watching: GPT-5 was originally expected much earlier this year, and its delays led some critics to argue that OpenAI's scaling-up strategy was yielding diminishing returns.
[28]
OpenAI launches GPT-5, nano, mini and Pro -- not AGI, but capable of generating 'software-on-demand'
Want smarter insights in your inbox? Sign up for our weekly newsletters to get only what matters to enterprise AI, data, and security leaders. Subscribe Now After literally years of hype and speculation, OpenAI has officially launched a new lineup of large language models (LLMs), all different-sized variants of GPT-5, the long-awaited predecessor to its GPT-4 model from March of 2023, nearly 2.5 years ago. The company is rolling out four distinct versions of the model -- GPT-5, GPT-5 Mini, GPT-5 Nano, and GPT-5 Pro -- to meet varying needs for speed, cost, and computational depth. * GPT-5 is the full-capability reasoning model, used in both ChatGPT and OpenAI's application programming interface (API) for high-quality general tasks * GPT-5 Pro is an enhanced version with extended reasoning and parallel compute at test time, designed for use in complex enterprise and research environments. It provides more detailed and reliable answers, especially in ambiguous or multi-step queries . * GPT-5 Mini is a smaller, faster version of the main model, optimized for lower latency and resource usage. It is used as a fallback when usage limits are reached or when minimal reasoning suffices. * GPT-5 Nano is the most lightweight variant, built for speed and efficiency in high-volume or cost-sensitive applications. It retains reasoning capability, but at a smaller scale, making it ideal for mobile, embedded, or latency-constrained deployments Like GPT-4 and its iterations launched in the preceding years and months -- GPT-4o, GPT-4.1, GPT-4.5 and others -- these four new GPT-5 LLMs will be used to power OpenAI's marquee product ChatGPT and serve answers to its 700 million weekly users. The model is now live within ChatGPT, where, as per rumors and reports, it replaces the previous complex system of model switching with an automatic router that finds the best variant of GPT-5 to answer a given user's unique input or query -- that is, more difficult queries will be routed to the GPT-5 pro and main models, while simpler queries will go towards GPT-5 mini and nano. In the API, the three reasoning-focused models -- GPT-5, GPT-5 mini, and GPT-5 nano -- are available as gpt-5, gpt-5-mini, and gpt-5-nano, respectively. GPT-5 Pro is not currently accessible via API, being used only to power ChatGPT. GPT-5's release comes just days after OpenAI launched a set of free, new open source LLMs under the name GPT-oss, which can be downloaded, customized and used offline by individuals and developers on consumer devices like PCs/Mac desktops and laptops. The biggest takeaway, though, is likely not what GPT-5 is, but what it isn't: AGI, artificial general intelligence, OpenAI's stated goal of an autonomous AI system that outperforms humans at most economically valuable work. Whether or not you the reader personally believe such a system is possible or desirable, OpenAI declaring AGI would have material business impacts. Wired reported previously that there is a clause in OpenAI's contract with Microsoft that permits OpenAI to begin charging Microsoft for access to its newest models, or cut it off from accessing OpenAI models, if OpenAI's board determines the company has achieved AGI or generates more than $100 billion in profit. But apparently, that is not the case today. As co-founder and CEO Sam Altman said, flanked by other OpenAI staffers on an embargoed video call with reporters last night, "the way that most of us define AGI, we're still missing something quite important -- many things that are quite important, actually -- but one big one is a model that continuously learns as its deployed, and GPT-5 does not." I also asked OpenAI the following question directly: "Is OpenAI considering GPT-5 AGI? Will it trigger any changes regarding Microsoft negotiations?" To which an OpenAI spokesperson responded over email: GPT-5 is a significant step toward AGI in that it shows substantial improvements in reasoning and generalization, bringing us closer to systems that can perform a wide range of tasks with human-level capability. However, AGI is still a weakly defined term and means different things to different people. While GPT-5 meets some early criteria for AGI, it doesn't yet reach the threshold of fully human-level AGI. There are still key limitations in areas like persistent memory, autonomy, and adaptability across tasks. Our focus remains on advancing these capabilities safely, rather than speculating on specific timelines. Why use GPT-5? With so many alternate models available now from OpenAI and a growing list of competitors, namely Chinese startups offering powerful open source models, what does GPT-5 bring to the table? Altman described the leap in capability as more than incremental. He compared the experience of using GPT-5 to upgrading from a pixelated display to a retina screen -- something users simply don't want to go back from. "GPT-3 felt like talking to a high school student," Altman said. "GPT-4 was like a college student. GPT-5 is the first time it feels like talking to a PhD-level expert in your pocket." Among the most impressive capabilities demoed for reporters during the embargoed call was the ability to generate the code for a fully working web application from a single prompt, in this case, a French language learning app with built-in game where English-to-French phrases were shown every time the user guided a virtual mouse to collect slices of cheese, with fully working emoji-inspired characters, backdrop/setting, and clickable interactive menus. The given prompt was only a single paragraph, too. As Altman stated: "This idea of software on demand will be a defining part of the new GPT-5 era." However, this basic capability -- prompt to working software -- has been available already from prior OpenAI models such as o3 and o4-mini, o4-high, and rival services like Anthropic's Claude Artifacts, which I (and many others) have used for many months to create interactive first-person and clickable games as well. The advantage GPT-5 seems to offer in making games, apps, and other software from prompts seems to be in speed -- it produced this demo app in a matter of mere minutes -- and completeness, with very few discernible bugs and a completely playable experience in "one-shot," or from a single prompt without back-and-forth conversation, as the developers like to say. Available to ChatGPT free users and all plans GPT-5 is not restricted to premium subscribers. OpenAI has made the model available across all ChatGPT tiers, including free users -- a deliberate move aligned with the company's mission to ensure broad benefits from AI. Free-tier users can access GPT-5 and GPT-5 Mini, with usage limits -- though exactly what those usage limits are remains undefined for now, and I'd guess will likely change on an irregular cadence depending on demand. Subscribers to the ChatGPT Plus ($20 per month) tier receive higher usage allowances, while subscribers to the ChatGPT Pro ($200 monthly), Team ($30 per month or $240 annually), and Enterprise (variable pricing depending on company size and usage) customers get unlimited or prioritized access. GPT-5 Pro will become available to Team, Enterprise, and EDU customers in the coming days. The new unified ChatGPT experience eliminates the need to select a model manually. Once users reach usage limits on GPT-5, the system automatically shifts to GPT-5 mini -- a more lightweight but still highly capable fallback. Improved metrics across the board, including 100% in AIME 2025 Math According to OpenAI, GPT-5 offers the most accurate, responsive, and context-aware AI system the company has ever shipped. It reduces hallucinations, handles multi-step reasoning more reliably, and generates better-quality code, content, and responses across diverse domains. The GPT-5 system delivers ~45% fewer factual errors than GPT-4o in real-world traffic, and up to ~80% fewer when using its "thinking" mode. This mode, which users can trigger by explicitly asking the model to take its time, enables more complex and robust responses -- powered by GPT-5 Pro in certain configurations. In tests, GPT-5 Pro sets new state-of-the-art scores on benchmarks like GPQA (88.4%), AIME 2025 math (100% when using Python to answer the questions), and HealthBench Hard (46.2%). Performance improvements show up across key academic and real-world benchmarks. In coding, GPT-5 sets new state-of-the-art results on SWE-Bench Verified (74.9%) and Aider Polyglot (88%). On writing tasks, GPT-5 adapts more smoothly to tone, context, and user intent. It is better at maintaining coherence, structuring information clearly, and completing complex writing assignments. The improvements are not just technical -- OpenAI's team emphasized how GPT-5 feels more natural and humanlike in conversation. Health-related use cases have also been enhanced. While OpenAI continues to caution that ChatGPT is not a replacement for medical professionals, GPT-5 is more proactive about flagging concerns, helping users interpret medical results, and guiding them through preparing for appointments or evaluating options. The system also adjusts answers based on user location, background knowledge, and context -- leading to safer and more personalized assistance. One of the most significant updates is in safe completions, a new system that helps GPT-5 avoid abrupt refusals or unsafe outputs. Instead of declining queries outright, GPT-5 aims to provide the most helpful response within its safety boundaries and explains when it cannot assist -- a change that dramatically reduces unnecessary denials while maintaining trustworthiness. New developer tools for using GPT-5 through the API GPT-5 is also a major upgrade for developers working on agentic systems and tool-assisted workflows. OpenAI has introduced a suite of developer-friendly controls in the GPT-5 API, including: * Free-form function calling - Tools can now accept raw strings such as SQL queries or shell commands, without requiring JSON structure. * Reasoning effort control - Developers can toggle between rapid responses and deeper analytical processing depending on the task. * Verbosity control - A new parameter allows users to select whether responses are brief, standard, or detailed. * Structured outputs with grammar constraints - Developers can now guide outputs using custom grammars or regular expressions. * Tool call preambles - GPT-5 can now explain its reasoning before using tools or making external requests. Developers can access GPT-5 through OpenAI's platform at standard pricing: $1.25 per million tokens for input, with 90% cache discounts, and $10 per million tokens for output. The context window now spans 256,000 tokens (about the length of a 600-800 page book of text) allowing GPT-5 to handle substantially larger documents and more extensive conversations than its predecessor, GPT-4 Turbo. For those who require even more, GPT-4.1 (which supports 1 million-token context windows) remains available. Early enterprise testers have high praise Several high-profile companies have already adopted GPT-5 in early trials. JetBrains is using it to power intelligent developer tools, and Notion has integrated GPT-5 to improve document generation and productivity workflows. At AI developer tool startup Cursor, co-founder and CEO Michael Truell said in a quote provided to reporters by OpenAI: "Our team has found GPT-5 to be remarkably intelligent, easy to steer, and even to have a personality we haven't seen in any other model. It not only catches tricky, deeply-hidden bugs but can also run long, multi-turn background agents to see complex tasks through to the finish -- the kinds of problems that used to leave other models stuck. It's become our daily driver for everything from scoping and planning PRs to completing end-to-end builds. " Other customers report major gains: GitLab cites a drop in tool call volume, GitHub notes improvements in reasoning across large codebases, and Uber is testing GPT-5 for real-time, domain-aware service applications. At Amgen, the model has already improved output quality and reduced ambiguity in scientific tasks. More updates still to come GPT-5's launch coincides with several new features coming now and soon to ChatGPT. Users can now personalize the interface with chat colors (with exclusive options for paid users) and experiment with preset personalities like Cynic, Robot, Listener, and Nerd -- designed to match different communication styles. ChatGPT will also soon support seamless integration with Gmail, Google Calendar, and Google Contacts. Once enabled, these services will be automatically referenced during chats, with no manual toggling required. These connectors launch for Pro subscribers next week, with broader availability to follow. A new Advanced Voice mode understands instructions better and allows users to adjust tone and delivery. Voice will be available across all user tiers and included in custom GPTs. In 30 days, OpenAI will retire the older "Standard Voice Mode" and fully transition to this unified experience. With safer design, more robust reasoning, expanded developer tooling, and broad user access, GPT-5 reflects a maturing AI ecosystem that's inching closer to real-world utility on a global scale. OpenAI's approach this time is less about flash and more about integration. GPT-5 isn't a separate offering that users have to seek out -- it's simply there, powering the tools millions already use, making them smarter and more capable and unlocking a whole new raft of use cases for developers.
[29]
GPT-5 is here - 5 things you need to know about OpenAI's 'most useful' model yet
OpenAI's much‑anticipated livestream reveal of GPT‑5 crammed a lot into about an hour and a half of announcements and demonstrations. CEO Sam Altman didn't show any of the fear he claimed to feel about the new AI model, just a lot of pride at what he and his team had accomplished. Much of the presentation showed off GPT-5's technical milestones and how they translate into powerful new and upgraded AI features for users. Altman also had some lighter fare to unveil for ChatGPT, offering more customization options and ways for users to link their existing online footprint to ChatGPT. Here are the five most notable pieces to emerge from GPT-5's debut. GPT-5 is the next iteration of OpenAI's models, bigger and more powerful, but not dissimilar in its basic form. Even so, GPT-5 is big and complex enough to reach a new level in how it seems to reason. Essentially, you no longer have to spoon-feed it context or restate complex prompts three times, or at least not nearly as often. Multifaceted questions like how changing interest rates might affect Gen Z homeownership trends in mixed markets might take several prompts refined multiple times to provide the answers you seek with earlier ChatGPT models, but GPT‑5 can unpack the whole thing. Based on the demonstrations, GPT‑5 seems to parse each part separately and stitch it together. And it flags when there's a gap in its knowledge, which is far better than confidently hallucinating. It applies that way of thinking to how it interacts with users, too. While obviously not 'thinking', it does appear to read between the lines well enough to reflect a user's mood and even adjust its response to an expressed emotion. The model is supposed to be particularly good with math and coding software. Good enough to handle the increasingly popular pursuit of 'vibe coding,' where you simply describe a feeling or a mood of a piece of software, and the AI produces the HTML, CSS, and JavaScript to match your design vision. GPT-5 comes in many sizes in addition to the standard version. There's also the smaller gpt-5-mini, and an even leaner gpt-5-nano, which lives solely in the API. The big news is that free ChatGPT users now get access to both GPT-5 and mini, while ChatGPT Plus subscribers enjoy higher usage limits across the board. If you're a Pro user paying $200 a month, you'll now get unlimited GPT-5 access, along with access to the more powerful gpt-5-pro model and gpt-5-thinking. These both take longer to provide answers, but come back with deeper, more thoughtful responses. There is no need to pick and choose yourself, either, though. ChatGPT now picks the right model automatically based on what you're asking and what plan you're on. ChatGPT has a default, pleasantly bland personality, but GPT‑5 is advanced enough to offer more variety in the tone and style of the AI chatbot. If you don't want the usual neutral mode, you can choose "Cynic" for sarcasm with your answer, "Listener" if you're venting and just need it to echo back understanding, "Nerd" for a side of geeky trivia, and "Robot" for the purely mechanical response. These personalities don't undercut the answers you get, but they do flavor the response. Therefore, you might get dry wit with productivity tips from the "Cynic" tone or gentle encouragement in your goals from "Listener." Additionally, the chatbot's appearance can now be altered with the new color themes. If you're a paid subscriber, you'll soon be able to adjust the look of ChatGPT instead of switching between the usual black or white. ChatGPT's Voice Mode is getting an audio glow-up of its own. OpenAI is rolling out a much-improved version that not only works with custom GPTs but also adapts its tone and speech style based on your instructions and overall vibe. You can ask it to be snappier, slower, warmer, or whatever else you want. For ChatGPT Plus users, voice responses are now nearly unlimited. Free users still get access, too, with a few hours a day to chat hands-free. To streamline things, the old Standard Voice Mode is being phased out entirely within 30 days. After that, everyone will be on the same upgraded experience. Next week, ChatGPT Pro users will be able to hook up their Gmail, Google Calendar, and Google Contacts directly to ChatGPT. That means no more switching tabs to check if you're free next Tuesday or digging through threads to find that one email you definitely forgot to reply to. Once connected, ChatGPT will pull in what it needs to help respond to your queries. OpenAI assured users that it will only pull in the minimum needed and only when it's helpful. You don't need to say "check my calendar" or "pull that contact." The AI will do so based on whether you request something that requires it, like scheduling a meeting. It will pick a time that works for you and write the email on your behalf. Other subscription tiers are scheduled to get access to the connections in the near future, so this won't be limited to Pro forever. All of these upgrades leveraging GPT-5 point to OpenAI's bigger plans to make its AI models an intimate part of your life, not just a tool you occasionally turn to and feel annoyed about having to carefully parse answers. Smarter reasoning means less cleanup for the user. Vibe coding shifts AI from merely aping code to interpreting your intended use with software. The personalities and colors make the AI feel like it's more unique to you, not just a one-size-fits-all tool, especially with the more realistic voice and access to your email account and calendar GPT‑5 mimics awareness better than any of its predecessors. That means it could blend into our routines and become as second nature to use as our smartphones. Or at least, that's what OpenAI and its investors likely hope to see happen.
[30]
OpenAI launches GPT-5 with free access, cuts prices
OpenAI unveiled GPT-5 on Thursday, its most advanced artificial intelligence model yet, with CEO Sam Altman promising it delivers "PhD-level expert" capabilities across any field. The launch represents the company's biggest upgrade since ChatGPT debuted in November 2022 and now serves 700 million weekly users. The new model will be available to both free and paid ChatGPT users immediately, marking the first time OpenAI has made its most powerful technology available without requiring a subscription. The company is also undercutting competitors with aggressive pricing for business customers, making GPT-5 significantly cheaper than rival models from Anthropic and matching Google's rates. The model shows particular strength in coding, with OpenAI claiming it can create "beautiful and responsive websites, apps, and games with an eye for aesthetic sensibility in just one prompt." The company demonstrated GPT-5 building everything from typing games to drum simulators from single prompts, with improved understanding of design elements like spacing and typography. Early reviews from business users and developers show promise for enterprise applications. Box CEO Aaron Levie posted on X that his company saw major improvements in extracting data from complex documents like contracts and legal files, with better accuracy and fewer errors than previous AI models. Other early testers found GPT-5 excels at going beyond basic requests. Ethan Mollick, a Wharton professor who writes about AI and has advanced access, wrote in his Substack that the AI often anticipates what users need and delivers more comprehensive results. When he asked for a simple building game, GPT-5 created a fully functional 3D city builder with features he never requested. However, some advanced users found the model more cautious than competitors. Kieran Klaassen, general manager at AI company Cora, said GPT-5 represents an improvement over existing models but "not a leap into the future," noting it works well for precise tasks but lacks the aggressive autonomous approach preferred by some programmers. OpenAI also announced new productivity features that will launch next week, including integration with Gmail and Google Calendar for paid users, allowing ChatGPT to access schedules and emails to provide personalized assistance. The company added customizable interface colors and personality settings to tailor the AI's responses. The launch caps a week of major AI developments that underscore the breakneck pace of innovation in the field. Earlier this week, rival Anthropic released Claude Opus 4.1, an updated version of its flagship model claiming improved coding and research capabilities. OpenAI also surprised the industry by launching GPT-oss, its first open-source AI model in over five years, allowing developers to run AI locally on personal computers. The open-source release marks a strategic shift for OpenAI, which hadn't published model code in over five years. Altman called GPT-oss "the best and most usable open model in the world" and said he expects it to accelerate innovation across the field. Meta CEO Mark Zuckerberg has also been aggressively recruiting AI talent, reportedly offering packages worth more than a billion dollars to poach researchers from OpenAI and other competitors. These developments come amid massive funding rounds fueling AI competition. OpenAI recently raised $8.3 billion at a $300 billion valuation, earlier this month. The company reported that its annual revenue has grown to $13 billion, up from $10 billion in June, with projections to exceed $20 billion by year-end. Anthropic is reportedly raising up to $5 billion at a $170 billion valuation, while OpenAI's infrastructure partnership with SoftBank and Oracle involves $500 billion in planned investments over four years. The launch reflects OpenAI's push to maintain its lead as AI development accelerates across the industry. The company's strategy of offering advanced capabilities for free while aggressively pricing its business services suggests it's prioritizing market dominance over immediate profits in what has become a winner-takes-all race for AI supremacy. "We think you will love using GPT-5 much more than any previous AI," Altman said during the announcement. "It is useful, it is smart, it is fast, and it's intuitive."
[31]
OpenAI releases ChatGPT-5 as AI race accelerates
OpenAI on Thursday released a keenly awaited new generation of its hallmark ChatGPT, touting "significant" advancements in artificial intelligence capabilities, as a global race over the technology accelerates. ChatGPT-5 is rolling out free to all users of the AI tool, which is used by nearly 700 million people weekly, OpenAI said in a briefing with journalists. Co-founder and chief executive Sam Altman touted this latest iteration as "clearly a model that is generally intelligent." "It is a significant step toward models that are really capable," he said. Altman cautioned that there is still work to be done to achieve the kind of artificial general intelligence (AGI) that thinks the way people do. "This is not a model that continuously learns as it is deployed from new things it finds, which is something that, to me, feels like it should be part of an AGI," Altman said. "But the level of capability here is a huge improvement." GPT-5 is particularly adept when it comes to AI acting as an "agent" independently tending to computer tasks, according to Michelle Pokrass of the development team. "GPT-3 felt to me like talking to a high school student -- ask a question, maybe you get a right answer, maybe you'll get something crazy," Altman said. "GPT-4 felt like you're talking to a college student; GPT five is the first time that it really feels like talking to a Ph.D.-level expert in any topic." Vibe coding Altman said he expects the ability to create software programs on demand -- so-called "vibe-coding" -- to be a "defining part of the new ChatGPT-5 era." As an example, OpenAI executives demonstrated the bot being asked to create an app for learning the French language. With fierce competition around the world over the technology, Altman said ChatGPT-5 led the pack in coding, writing, health care and much more. Rivals including Google and Microsoft have been pumping billions of dollars into developing AI systems. Altman said there were "orders of magnitude more gains" to come on the path toward AGI. "Obviously...you have to invest in compute (power) at an eye watering rate to get that, but we intend to keep doing it." ChatGPT-5 was also trained to be trustworthy and stick to providing answers as helpful as possible without aiding a seemingly harmful mission, according to OpenAI safety research lead Alex Beutel. "We built evaluations to measure the prevalence of deception and trained the model to be honest," Beutel said. ChatGPT-5 is trained to generate "safe completions," sticking to high-level information that can't be used to cause harm, according to Beutel. The debut comes a day after OpenAI said it was allowing the US government to use a version of ChatGPT designed for businesses for a year for just $1. Federal workers in the executive branch will have access to ChatGPT Enterprise essentially free in a partnership with the US General Services Administration, according to the artificial intelligence sector star. The company this week also released two new AI models that can be downloaded for free and altered by users, to challenge similar offerings by US and Chinese competition. The release of gpt-oss-120b and gpt-oss-20b "open-weight language models" comes as the ChatGPT-maker is under pressure to share inner workings of its software in the spirit of its origin as a nonprofit.
[32]
Chat GPT 5.0 will dramatically change the way you use AI
OpenAI's much-anticipated ChatGPT 5.0 is expected to arrive late August, 2025 -- and it could change how everyday users interact with AI. While the leap from GPT 4 to GPT 5 may not be as dramatic as the jump from GPT 3, the improvements on the horizon could make AI feel less like a tool and more like a true assistant. The challenge for many users is how to efficiently tap into the power of Artificial Intelligence without having that become their full-time job. Chat GPT 5.0 should help to resolve that problem. When Is It Coming? Industry reports and hints from OpenAI CEO Sam Altman suggest a mid-to-late August 2025 launch. The rollout may be gradual, with early capacity crunches as millions rush to try it out. OpenAI has also flagged that the model has been undergoing extensive safety testing and external "red‑teaming" to ensure it's ready for mainstream use. Currently there is a version 4.5 model available in research mode. Why It Matters: Unified AI Model: No more switching between "reasoning" and "creative" modes -- GPT 5 blends them into one smarter assistant. Better Reasoning: Handles logic-heavy and multi-step problems more reliably, with fewer errors. Full Multimodality: Text, images, voice, and now video -- you can talk to GPT 5 in almost any medium. Massive Memory: Potentially up to 1 million tokens of context, letting it keep track of long conversations or entire documents. Smarter AI Agents: More capable at planning trips, booking appointments, or shopping online without constant supervision. Speed Options: Three tiers -- flagship, "mini," and "nano" -- so you can match performance to your needs and budget. Why Should I care? Your Conversations Will Flow Better: Longer memory means GPT‑5 won't "forget" what you said halfway through a project. It can remember key details from earlier in the conversation, making it feel more like talking to a person who's actually paying attention. This is an advantage versus Google Gemini which does not have the ability to track conversational threads in its commonly used models. You Can Work Across Formats Seamlessly: Need to summarize a meeting video, write a blog post, and design an image for social media? You'll be able to do it all in one conversation without juggling multiple tools. More Accurate, Less Frustrating Results: GPT‑5's reasoning boost should reduce the number of "hallucinated" facts, meaning you'll spend less time double-checking its work. If it delivers on this, this will be a major advance. Currently user need to give back some of the time that they save with AI to check its work. Smarter Help Without Micro-Managing: Imagine telling GPT‑5, "Plan me a weekend trip to Seattle," and it not only finds the flights but also suggests restaurants, books a hotel, and emails you an itinerary -- all without extra back-and-forth. Why the Excitement Is Justified Recommended Videos Sam Altman recently shared that GPT 5 solved a problem he couldn't -- prompting him to joke that he felt "useless" next to it. That level of problem-solving power is what's drawing attention. And for casual users, the big win isn't just in speed or smarts -- it's in making everyday tasks simpler, faster, and more fun. Ok, What's Next? ChatGPT‑5.0 is shaping up to be the most versatile and capable AI model yet. Whether you're looking to brainstorm creative ideas, get reliable answers, or simply save time on your to-do list, GPT‑5 promises to meet you where you are -- across text, voice, images, and video.If this release lives up to the hype, AI could move from being a novelty to becoming a daily habit. And we'll be here to cover every step of that journey.
[33]
ChatGPT-5 launch live: follow the build-up to OpenAI's major livestream event
OpenAI is expected to announce the latest model of its popular AI The much-anticipated of arrival of ChatGPT-5, the next major upgrade to OpenAI's ChatGPT AI model, looks like it will happen today. OpenAI posted a not-so-cryptic message on X, hinting at the new model's arrival with the number five replacing the 's' in livestream, while also confirming the time and date for the livestream launch. Recommended Videos ChatGPT-5 is set tipped to dramatically change the way you use AI, with a host of new features expected to be announced. When is the ChatGPT-5 launch livestream? The ChatGPT-5 launch event will start at 10am PT / 1pm ET today. How to follow the ChatGPT-5 launch live We will be running our ChatGPT-5 live blog during the build-up to the event, and throughout the launch livestream itself, keeping you in the know every step of the way. OpenAI hasn't revealed details on exactly where you'll be able to watch the launch event just yet, but we'd recommend keeping a close eye on the following places; The OpenAI news site OpenAI's X feed OpenAI's YouTube channel ChatGPT-5 launch live blog LiveLast updated August 07, 2025 4:27 AMThe liveblog has ended.No liveblog updates yet. Load more
[34]
OpenAI launches GPT-5, its most powerful AI yet -- will it be enough to stay ahead in today's ruthless AI race?
But 2025 has become a ruthless race for AI dominance, and OpenAI has struggled to remain the undisputed pace-setter against a growing field of rivals developing advanced LLM models. On Thursday, OpenAI took a major step in its effort to reassert its leadership with the launch of GPT-5, the long awaited update to its flagship AI product and its most powerful and fastest model yet. The company said the model delivers "more accurate answers than any previous reasoning model," and is "much smarter across the board," reflected by strong performance on academic and human-evaluated benchmarks. Its research blog noted new state-of-the-art performance across math, coding and health questions, and found that GPT-5 outperformed other OpenAI models across tasks spanning over 40 occupations including law, logistics, sales and engineering. In addition, it is being billed as "one unified system" that provides "the best answer, every time," with no need to pick from what was becoming a laundry list of different OpenAI models. "GPT-5 really feels like talking to a PhD level expert in any topic," OpenAI CEO Sam Altman told journalists in a pre-briefing on Wednesday. "Something like GPT-5 would be pretty much unimaginable in any other time in history." Altman described GPT-5 as a "significant step" along the path to artificial general intelligence (AGI), which according to OpenAI's mission statement is defined as "highly autonomous systems that outperform humans at most economically valuable work." OpenAI is making its latest AI model free to all ChatGPT users -- the first time free users will have access to one of its reasoning models -- as well as through an API that lets developers and businesses build on top of it. OpenAI is also rolling out some new ChatGPT features: Users can choose from four pre-set personalities -- Cynic, Robot, Listener, and Nerd -- to customize how the AI responds, while Pro users will soon be able to connect Gmail, Google Calendar, and Contacts, allowing ChatGPT to reference that information automatically during chats. Voice mode is also getting an upgrade, with more adaptive and expressive responses. It's unclear whether this combination of speed, power and features will be enough, however. Some two years in the making (GPT-4 was launched in March 2023), GPT-5's launch has taken longer than many industry insiders expected, as OpenAI has adjusted its approach in response to industry changes. And while ChatGPT now boasts an impressive 700 million weekly users, OpenAI has faced growing pressure over the past year as rivals poach its talent and race ahead on emerging AI techniques like long-context reasoning and autonomous tool use. In addition to Big Tech competitors like Meta and Google, OpenAI must contend with a wave of startups founded by its own former researchers, including Anthropic, Thinking Machines, and Safe Superintelligence. Mark Zuckerberg's Meta has emerged as a particularly aggressive rival, forming a new Superintelligence team that has lured away several top OpenAI scientists. And in January, Chinese upstart DeepSeek briefly knocked OpenAI back on its heels -- part of a growing flood of powerful Chinese models now vying for global influence. Whether GPT-5 propels OpenAI back to the top of the AI hill will become clear in the days and weeks ahead, as researchers put the model through its paces, testing it against the likes of other elite models, including Anthropic's latest Claude model and Google's Gemini. One of the defining truths about the world of generative AI is that even when you're on top, the lead doesn't last for long. Now that GPT-5 is out, OpenAI CEO Sam Altman acknowledged that staying at the frontier means one thing: relentless scaling. In AI, scaling refers to the idea that models get more powerful as you increase the amount of data, computing power, and model components used during training. It's the underlying principle that drove progress from GPT-2 to GPT-3 to GPT-4 -- and now GPT-5. The catch is that each leap requires exponentially more investment, particularly in AI infrastructure -- for OpenAI, that includes its Stargate Project, a joint venture it announced in January with Softbank, Oracle and investment firm MGX with a goal to to invest up to $500 billion by 2029 in AI-specific data centers across the U.S. When asked whether scaling laws still hold, Altman said they "absolutely" do. He pointed to better models, smarter architectures, higher-quality data, and significantly more computing power as the path to "order-of-magnitude" improvements still ahead. But that kind of progress comes at a cost. "It's going to take an eyewatering amount of compute," he admitted. "But we intend to continue doing it." OpenAI has roughly doubled its revenue in the first seven months of 2025, hitting an annualized run rate of $12 billion -- up from about $6 billion at the start of the year, according to a recent report by The Information. That translates to $1 billion in monthly revenue, fueled by surging demand for its ChatGPT products across both consumer and enterprise markets. Weekly active users for ChatGPT have jumped to around 700 million, up from 500 million across all OpenAI products as of late March. And earlier this week OpenAI released a free, open-source model -- an unusual move for a company often criticized for its closed approach over the past half-decade -- suggesting confidence that its premium offering, which is now GPT-5, will continue to dominate. There are plenty of challenges ahead, however. For one thing, the partnership between Microsoft and OpenAI -- that began with a $1 billion investment in 2019 -- is entering a more fraught and complex phase. While Microsoft has invested more than $13 billion and retains exclusive rights to OpenAI's models through Azure, tensions have emerged over revenue sharing, AGI control clauses, and overlapping product strategies. OpenAI is also navigating an effort to turn its commercial arm into a Public Benefit Corporation (PBC) while ensuring its original nonprofit maintains control. There has been significant legal and public backlash to its efforts, including a lawsuit from co-founder Elon Musk and scrutiny from state attorneys general in California and Delaware. In addition, OpenAI faces broader regulatory attention as it rethinks its governance structure -- raising questions about charitable asset protection, public benefit accountability, and compliance with state nonprofit laws.
[35]
OpenAI Releases GPT-5, Says It's Shutting Down All Previous Models
OpenAI has released its long-awaited GPT-5 AI model, making it free to use for anybody as of today. The Sam Altman-led company calls it "our smartest, fastest, and most useful model yet," making it "more useful responses across math, science, finance, law, and more." "GPT-5 is the first time that it really feels like you're talking to an expert in any topic," Altman said during a press briefing today, likening it to Apple's iPhone switching to high-density Retina displays, whose individual pixels could no longer be made out by the naked eye. Altman didn't hold back, asserting that it's the "best model in the world at coding" and "writing," as well as "health care, and a long list of things beyond that." GPT-5 is a "reasoning" model, which means that it takes an iterative approach to problem solving before coming up with its final output. The company is putting its money where its mouth is. "With GPT-5, we're deprecating all of our old models," said OpenAI researcher Tina Kim during the event, using a term the company uses when it's shutting down an obsolete model. The company also claims that GPT-5 addresses some major pain points with the tech, including hallucinations, which continue to haunt pretty much all currently available AI models to this day, as well as cybersecurity. But how all of this will shake out during real-world use remains to be seen. While OpenAI claims GPT-5 will make it easier than ever before to link one's Google account, Wired's recent reporting on glaring cybersecurity gaps plaguing ChatGPT should make any user think twice about using the company's Connectors feature. The company also claims that GPT-5 is more robust to hijacking attempts -- but we'll wait until hackers have had enough time with the tool to draw any conclusions. OpenAI is aiming to enhance customizability by offering four preset personalities, including "cynic," "robot," "listener," and "nerd." In spite of all the braggadocio, Altman admitted that OpenAI was still a very long way from fulfilling its core goal of achieving "artificial general intelligence" (AGI) -- a poorly defined term that roughly denotes the point at which an AI's capabilities surpass those of a human -- that would benefit all of humanity. "I kind of hate the term AGI, because everyone at this point uses it to mean a slightly different thing, but [GPT-5] is a significant step forward towards models that are really capable," Altman said today. "We're still missing something quite important." It also remains to be seen whether OpenAI has meaningfully addressed the issue of users suffering delusions in the form of "ChatGPT psychosis" with its latest model. Prior to today's announcement, the company released an update this week, announcing that it will be including lackluster "gentle reminders during long sessions to encourage breaks." In short, GPT-5 appears to be yet another incremental step forward as the AI industry attempts to justify the tens of billions of dollars being poured into infrastructure buildouts to power incredibly environmentally damaging AI models. The company already rushed out two "state-of-the-art open-weight language models," dubbed "gpt-oss-120b and gpt-oss-20b," earlier this week, though they're certain to now be overshadowed by GPT-5. One thing's for sure: the former nonprofit is red hot right now. On Wednesday, news emerged that the company is deep into talks about a potential secondary stock sale that would value it at roughly half a trillion dollars.
[36]
After Months of Speculation, OpenAI Finally Launches GPT-5 - Decrypt
The company released two open-weight models earlier this week. After weeks of speculation, OpenAI unveiled the next generation of its flagship AI model, GPT‑5. The new model, which creates a simpler interface to the world's most popular AI platform, arrives amid mounting pressure from rivals like Meta, Anthropic, and Google. "GPT is a significant upgrade over GPT-4 and a significant step along the path to AGI," CEO Sam Altman said, noting that among other things, GPT-5 can "write a software program from scratch. We think this idea of software on demand will be a defining characteristic of the GPT-5 era." The new model wrote over 400 lines of code in two minutes, in a "zero shot" example from a prompt asking it to do a simulation of the Bernoulli effect. The new model, which offers a video option in which it can watch you (or anything else via your camera) while you chat, will be available to all GPT users, including the free tier, starting today. Free users will be capped on use and then be rolled over to a "mini" model. Altman positioned GPT-5 as a qualitative leap in AI capability during Thursday's livestream. "GPT-4 is like talking to a college student. Now, GPT-5 is like talking to an expert," he said, while talking about the company's ambition to move beyond incremental improvements toward fundamentally different AI interactions. Business Integration Focus The company emphasized enterprise applications throughout the presentation. "It will be an especially important moment for businesses," Altman noted, highlighting GPT-5's enhanced reasoning capabilities as a potential transformer for corporate workflows and decision-making processes.Unified Reasoning Architecture. OpenAI described reasoning as central to its artificial general intelligence strategy. "Reasoning is at the heart of our AGI program," the team explained. The breakthrough eliminates previous trade-offs between speed and analytical depth: "Until now users had to pick between slow answers using reasoning and fast answers with normal token prediction. This model eliminates that barrier." Enhanced Performance Metrics The company touted comprehensive improvements across key metrics. "GPT-5 is faster, more reliable and more accurate" than its predecessor, according to internal testing. OpenAI also positioned the model as "the best coding model to date," with demonstrated software development capabilities that surpass previous iterations.The model also beats all the scores in different benchmarks shared by the company, especially in math and coding. Access Structure OpenAI announced a tiered rollout strategy for the new model. Free tier users will start with GPT-5 before transitioning to a lighter "GPT-5 mini" version when they deplete their usage quota, while Pro subscribers receive unlimited access to the full model. The company confirmed that all existing ChatGPT features will remain compatible with the new system. The company is deprecating its older models, and asked GPT-5 to write an obituary for them, which one of the presenters pointed out was more natural in its reponse, like talking to someone with a higher IQ and higher EQ she said: "It has more rhythm ... It's actually quite personal -- and that's the kind of thing GPT-5 does better... the responses feel less like AI." The model also has native multilingual capabilities, making it more capable than other models in multilingual tasks ChatGPT currently boasts 700 million weekly active users, compared to Google Gemini's 47 million weekly users. Despite its large user base, GPT-5 will be competing with Meta AI, whose own weekly user base stands at 700 million, thanks to its integration with Instagram and WhatsApp. To date, OpenAI has raised $57 billion across 11 funding rounds, according to data from Tracxn, and is valued at $300 billion. According to a Reuters report, that valuation number is expected to rise to $500 billion if a proposed stock sale option goes through. The company doubled its revenue in the first seven months of the year and is expected to reach $20 billion by December 31, Reuters reported. Altman first publicly confirmed that OpenAI started working on GPT‑5 in 2023, with the actual training phase beginning one year later. On February 13, 2025, Altman outlined the company's roadmap and said GPT‑5 was expected to launch "within months," with the aim of unifying OpenAI's O‑Series and GPT models. Altman has recently spent a great deal of time opining about artificial general intelligence, fueling speculation that OpenAI could use today's event to announce a breakthrough. Earlier this month, however, he cautioned that there could be additional delays in the rollout of GPT, asking users to "please bear with us." In the run-up to the launch earlier this week, OpenAI released two open-weight models on Tuesday -- gpt‑oss‑120b and gpt‑oss‑20b -- under an open license, aimed at developers seeking to run models locally.
[37]
Sam Altman's OpenAI Confirms GPT-5 Is About to Drop - Decrypt
Github leaks point to different GPT-5 variants touting agentic-like logic and unified architecture, with leaks benchmarks showing outstanding scores OpenAI dropped a teaser tweet today featuring a minimalist 7-second video where the text "GPT-4o" repeatedly appears on a white background before gradually transitioning -- through subtle fades and accents -- to "GPT-5." The post, captioned simply "Dropping soon," garnered over 11,000 likes, and has been watched over half a million times within hours. GitHub leaks that surfaced earlier this week revealed four variants in development: a base GPT-5 for logic and multi-step tasks, gpt-5-mini for efficiency, gpt-5-nano for low-latency applications, and gpt-5-chat tailored for enterprise conversations. The leaks mentioned features like "enhanced agentic capabilities" for handling complex coding with minimal prompting -- essentially an AI that takes initiative rather than waiting for detailed instructions. The base GPT-5 supposedly handles multi-step logic and autonomous decision-making with minimal user input, acting more like what the leaks called an "initiative-taking intelligence." The model, OpenAI CEO Sam Altman said months ago, is expected to unify the reasoning process and the normal token prediction in one powerful, hybrid model capable of merging what today are the "GPT" and the "o" architectures into one. The unified architecture represents a significant shift from OpenAI's current approach. Instead of users manually selecting between different models, GPT-5 will automatically route queries to the appropriate internal model version or research method. The system combines text, vision, and audio processing in a single neural network, rather than the current pipeline of separate models. And if we're to trust the leaks, the benchmarks look pretty interesting, with GPT-5 being by far the most powerful AI model in different tasks. OpenAI's head of applied research expressed excitement yesterday about seeing "how the public responds to GPT-5," while warning that infrastructure scaling has caused delays from earlier expectations. The company released two open-weight models, gpt-oss-120b and gpt-oss-20b, earlier this week -- the latter capable of running locally on consumer PCs -- potentially as a precursor to the main event. The reactions ranged from pure hype to healthy skepticism. Some users posted memes of eager anticipation with comments begging OpenAI to "drop it" already, while others worried about access tiers. "GPT-5 will probably only be available to high tier subscriptions. What's the point when we can't even use?" one frustrated user wrote, with others posting copium screenshots of the model being available for Plus and Free tier users. For the doubters still clinging to ambiguity, we decided to ask ChatGPT directly what the teaser meant. The bot didn't mince words: "The transition from GPT-4o to GPT-5 in the video clearly indicates the announcement of the next-generation model." When pressed further about whether this was definitely GPT-5, it added that the visual progression "leaves little room for alternative interpretation."
[38]
The GPT-5 Cheat Sheet: 13 Things to Know About OpenAI's Latest AI Leap - Decrypt
OpenAI touts "expert-level reasoning" and a memory that never sleeps -- plus major upgrades to coding, creative writing, and reliability. OpenAI unveiled GPT-5 during a Thursday livestream, marking what the company called a qualitative shift in artificial intelligence capability after several months of anticipation and multiple delays. The model is being rolled out to all ChatGPT users throughout the day today. The release represents OpenAI's attempt to unify its various AI technologies into a single system. The company described reasoning as central to its artificial general intelligence strategy, with the breakthrough eliminating previous trade-offs between speed and analytical depth. Users no longer need to choose between fast responses and deep reasoning capabilities -- GPT-5 delivers both simultaneously. GPT-5 rolls out today on ChatGPT and via its API. Microsoft has also incorporated GPT-5 into its products immediately, making it available through Copilot and GitHub Copilot. Yes, sort of: Free tier users will start with the standard GPT-5 before transitioning to a lighter "GPT-5 mini" version when they deplete their usage quota. Pro subscribers ($200 a month) get unlimited access to the full model while Plus subscribers ($20/month) get access to standard GPT-5. Pro subscribers can run GPT-5 at its highest intelligence level with additional features like early access to its advanced agents, unlimited usage, more capabilities for deep research, priority access, and advanced voice mode with higher limits for video and screen sharing. 3. What does multimodal mean? Does a separate image generator go away? Multimodal means GPT-5 can process and generate different types of content -- text, images, voice, and now even video -- all within the same conversation. The model showed enhanced foreign language understanding for complex tasks, generating complete websites with French words and proper pronunciation. Instead of juggling between Vision, Sora, GPT, and the "o" models to reason, GPT-5 can do everything on its own. 4. How big is the context window and why does it matter? GPT-5 has a 256,000 token context window for input, with the API accepting up to 272,000 input tokens and emitting a maximum of 128,000 reasoning and output tokens, for a total context length of 400,000 tokens. This means it can process roughly 200,000 words at once -- equivalent to a long novel. The larger context window allows GPT-5 to maintain coherent conversations over much longer interactions and analyze entire codebases or lengthy documents without losing track of important details. That said, this window is not very big by today's standards. Just for context, Gemini 2.5 is capable of handling 1 million tokens, 5. What new features does it have? None, really, but some of its skills are upgraded to such a degree that they will feel like new features. 6. So what's so great about it? GPT-5 is more powerful in just about every way. For instance, it demonstrated remarkable coding capabilities during the presentation, writing over 400 lines of code in two minutes when prompted to create a Bernoulli effect simulation from scratch. Other cool things shown off in the demo: 7. Has pricing changed? ChatGPT subscription pricing remains unchanged at $20/month for Plus and $200/month for Pro. For API users, GPT-5 costs $1.25 per million input tokens and $10.00 per million output tokens for the standard model. GPT-5 mini costs $0.25 per million input tokens and $2.00 per million output tokens, while GPT-5 nano runs $0.05 for input and $0.40 for output. This would make the model competitive against offerings from other companies and even cheaper than other models from OpenAI like GPT-4.1 or OpenAI o1 pro which costs a whopping $600 per million tokens. 8. Are we at AGI yet? No. However, the company positioned reasoning as "at the heart of our AGI program." The model represents significant progress but remains focused on specific tasks rather than matching human intelligence across all domains. For instance, GPT-5 is great at language tasks but lacks the general intelligence required to perform a wide range of activities independently. It's not yet self-teaching or self-adapting. 9. Can GPT-5 generate videos? Not yet. While video generation wasn't included in the initial release, OpenAI has Sora for video creation as a separate product. CEO Sam Altman previously indicated that future versions would support video "eventually." The current version does understand live video, however, so it could watch you try to fix a bike and provide live instructions. 10. How reliable is it compared to previous models? OpenAI reported that GPT-5 is "significantly less deceptive" than previous models, addressing one of the most persistent challenges in large language model deployment. On factual accuracy benchmarks, GPT-5 makes approximately 80% fewer factual errors than o3, making it substantially more trustworthy for enterprise applications according to Jakub Pachocki, OpenAI's chief scientist. 11. What about memory and personalization? GPT-5 supposedly will offer better persistent memory across sessions, remembering facts, preferences, and instructions across multiple conversations, even if you close the app and open a new tab days later. GPT-4's memory was limited, especially days after a session paused. The company said you can now set long-term objectives (e.g. help me lose 10 pounds in a healthy way, or help me prepare for my physics test), and GPT-5 will adapt its responses accordingly to proactively align with your goals. 12. How private is my personal data? Altman previously acknowledged that OpenAI might have to hand over a user's personal data to the government if legally required to do so. 13. Do I need to switch between different models anymore? Not anymore -- unless you want to generate video via Sora. With GPT-5's launch, OpenAI expressed confidence in deprecating all previous models. The company designed GPT-5 to handle all use cases that previously required specialized models, though users can still choose between GPT-5, GPT-5 mini, and GPT-5 nano based on speed and cost requirements.
[39]
OpenAI releases ChatGPT-5 as AI race accelerates
San Francisco (United States) (AFP) - OpenAI on Thursday released a keenly awaited new generation of its hallmark ChatGPT, touting "significant" advancements in artificial intelligence capabilities, as a global race over the technology accelerates. ChatGPT-5 is rolling out free to all users of the AI tool, which is used by nearly 700 million people weekly, OpenAI said in a briefing with journalists. Co-founder and chief executive Sam Altman touted this latest iteration as "clearly a model that is generally intelligent." "It is a significant step toward models that are really capable," he said. Altman cautioned that there is still work to be done to achieve the kind of artificial general intelligence (AGI) that thinks the way people do. "This is not a model that continuously learns as it is deployed from new things it finds, which is something that, to me, feels like it should be part of an AGI," Altman said. "But the level of capability here is a huge improvement." GPT-5 is particularly adept when it comes to AI acting as an "agent" independently tending to computer tasks, according to Michelle Pokrass of the development team. "GPT-3 felt to me like talking to a high school student -- ask a question, maybe you get a right answer, maybe you'll get something crazy," Altman said. "GPT-4 felt like you're talking to a college student; GPT five is the first time that it really feels like talking to a PhD-level expert in any topic." Vibe coding Altman said he expects the ability to create software programs on demand -- so-called "vibe-coding" -- to be a "defining part of the new ChatGPT-5 era." As an example, OpenAI executives demonstrated the bot being asked to create an app for learning the French language. With fierce competition around the world over the technology, Altman said ChatGPT-5 led the pack in coding, writing, health care and much more. Rivals including Google and Microsoft have been pumping billions of dollars into developing AI systems. Altman said there were "orders of magnitude more gains" to come on the path toward AGI. "Obviously...you have to invest in compute (power) at an eye watering rate to get that, but we intend to keep doing it." ChatGPT-5 was also trained to be trustworthy and stick to providing answers as helpful as possible without aiding a seemingly harmful mission, according to OpenAI safety research lead Alex Beutel. "We built evaluations to measure the prevalence of deception and trained the model to be honest," Beutel said. ChatGPT-5 is trained to generate "safe completions," sticking to high-level information that can't be used to cause harm, according to Beutel. The debut comes a day after OpenAI said it was allowing the US government to use a version of ChatGPT designed for businesses for a year for just $1. Federal workers in the executive branch will have access to ChatGPT Enterprise essentially free in a partnership with the US General Services Administration, according to the artificial intelligence sector star. The company this week also released two new AI models that can be downloaded for free and altered by users, to challenge similar offerings by US and Chinese competition. The release of gpt-oss-120b and gpt-oss-20b "open-weight language models" comes as the ChatGPT-maker is under pressure to share inner workings of its software in the spirit of its origin as a nonprofit.
[40]
Everything We Know About GPT-5, OpenAI's Latest Model
The post reveals the new model improves "reasoning, code quality, and user experience," with advancements in coding tasks as well as "enhanced agentic capabilities." OpenAI's upcoming model, GPT-5, is nearly here. The company plans on announcing the model in a "longer than usual" presentation today, which suggests the company has some interesting news up its sleeve. The thing is, the cat's out of the bag -- at least partially -- as some GPT-5 news has leaked ahead of the announcement. In a now-deleted blog post (archived here), we can see some of the new features and changes shipping with GPT-5. It'll come as no surprise that the post claims GPT-5 is OpenAI's "most advanced model," nor will it be a shock to learn that it improves "reasoning, code quality, and user experience" over previous models. However, the post does say the model can create and run code with "minimal prompting," and that it now has "enhanced agentic capabilities." Companies are now pushing agentic AI more than ever: Rather than simply generate results, like text, images, or videos, from prompts, agentic AI can run tasks on the users' behalf. ChatGPT's agent mode, for example, can take over your web browser to perform tasks for you, or create an excel doc based on a request for a specific spreadsheet. The post doesn't offer many other details in regard to GPT-5's performance and feature gains over previous models. We don't have any charts or figures to work from at this time. However, the post does share that GPT-5 is broken into four distinct models, depending on the use case. That includes: In addition to this leaked GitHub post, BleepingComputer reports it has seen documents suggesting OpenAI's pricing plan for GPT-5. Like previous models, users will be able to access GPT-5 for free. However, ChatGPT Plus subscribers may be able to access GPT-5 with "advanced reasoning," while ChatGPT Pro users may get "pro reasoning." Reasoning models take extra time to "think" through a query, breaking down the request into individual steps before processing the request. BleepingComputer says GPT-5 can sync between reasoning and non-reasoning automatically, but that you can prompt it to reason at any time. That's about the extent of our GPT-5 knowledge as of this article. However, we won't have to wait long to learn more: The leak comes just hours before OpenAI is set to officially announce its latest model, so it won't be long before we see how GPT-5 compares to other models, like GPT-4o, Google's Gemini, and Anthropic's Claude. You can check out the announcement live today at 10 a.m. PT (1 p.m. ET). To watch, head to OpenAI's official YouTube channel, and click on the GPT-5 livestream.
[41]
OpenAI unveils ChatGPT‑5. Here's what to know about the latest version of the AI-powered chatbot.
Anne Marie D. Lee is an editor for CBS MoneyWatch. She writes about topics including personal finance, the workplace, travel and social media. A new version of ChatGPT has arrived that OpenAI CEO Sam Altman promises will have Ph.D-level smarts. OpenAI on Thursday announced the release of GPT‑5, which it calls its "smartest, fastest and most useful model yet." The artificial intelligence company that brought the world ChatGPT says its latest version of the AI-powered chatbot will be more accurate, have fewer hallucinations and offer more articulate writing capabilities for composing emails and reports, for example. ChatGPT-5 will also excel at coding and answering health-related questions. A basic version of the new model is available for free, with paid options for higher usage also available. On a call with reporters Wednesday for a preview of the new chatbot, Altman likened ChatGPT-5 to a Ph.D.-level expert. The new chatbot is also "the biggest single step forward" that OpenAI has taken in worldwide accessibility, Altman said. The announcement marks the next step in AI development for OpenAI, which launched the first iteration of ChatGPT in 2022. The technology quickly captured the fascination of the tech industry and the public for its ability to generate human-like responses to questions and requests. OpenAI has since launched a series of updates to its chatbot, which now has over 700 million weekly users, according to the company. Read on to learn more about the latest version of ChatGPT. ChatGPT‑5 will offer more accurate responses in a shorter time frame than previous models, executives said on Wednesday's call. "You really get the best of both worlds" Nick Turley, head of product at ChatGPT, told reporters on the call. "You have it reason when it needs to reason, but you don't have to wait as long." The new version is also the best at coding to date, allowing users to build websites from scratch within minutes. In a demo during the call, Altman used GPT-5 to create another large language model, or GPT, in less than 5 minutes. Altman called the new chatbot's ability to write code on demand its "superpower," adding that the advancement would have been "unimaginable at any previous point in history." Asked about how the technology might impact the livelihood of human programmers, Altman said he thought the technology would actually create more job opportunities for as demand for software rises. According to OpenAI, the new model will also be better at answering health-related questions, flagging potential concerns and helping users understand test results from their doctor. The company noted, however, that the technology "does not replace a medical professional." Users around the world will be able to access ChatGPT-5 for free, according to OpenAI. Asked about the commercial rationale behind offering a free global model, Turley said the company's mission is to ensure AI benefits all humanity. "Giving everyone access to this capability is a very concrete way for us to live and breathe that mission," he said. In addition to the free version, OpenAI will also offer a variety of paid options based on usage limits. The subscription models include:
[42]
OpenAI releases GPT-5, calling it a 'team of Ph.D. level experts in your pocket'
Open AI CEO Sam Altman, at the Snowflake Summit on June 2, in San Francisco.Justin Sullivan / Getty Images OpenAI on Thursday released the newest version of the artificial intelligence model that powers its popular ChatGPT chatbot, with CEO Sam Altman promoting it as like having a "team of Ph.D. level experts in your pocket." The company said the new model, GPT-5, is its smartest and fastest to date with wide-ranging improvements to ChatGPT's skills in areas like coding, writing and taking on complex actions. The model has been eagerly anticipated by technologists as a leap forward in AI capabilities and a step toward the kind of superintelligent systems that almost every major technology company is now racing to build. OpenAI along with Microsoft, Meta, Google, Amazon and others have already plowed hundreds of billions of dollars into AI investment, with some projections of future spending reaching into the trillions. But GPT-5's new abilities appear to be largely refinements to features that ChatGPT and other AI systems already have. During a briefing Wednesday with the news media, OpenAI did not share any benchmark evaluations or tests that would rank GPT-5's abilities against those of other models -- as is common in many AI product releases. A spokesperson told NBC News that there are still "key limitations in areas like persistent memory, autonomy, and adaptability across tasks." GPT-5 is free for everyone. But users in paid tiers will get higher usage limits, as well as access to GPT-5 Pro. Altman said he believes the new model is so smart that going back to its predecessor GPT-4, which launched in 2023, feels "quite miserable" in comparison. "It reminds me of when the iPhone went from those giant-pixel old ones to the retina display, and then I went back to using one of those big pixelated things," Altman said. "And I was like, 'Wow, I can't believe how bad we had it.'" Starting Thursday, ChatGPT users will also be able to choose a custom color for chats, pick from a selection of pre-set personalities and connect their ChatGPT to Gmail, Google Calendar and Google Contacts. GPT-5 comes as OpenAI looks to find an edge in the global AI race that has pushed tech companies to roll out sweeping advancements in recent years, with full support from their respective governments. As generative and "agentic" (AI that can autonomously take actions on a user's behalf) AI tools grow ever more sophisticated, they've been met with a mix of excitement and ethical concerns across industries. The world has seen rapid technological advancements since ChatGPT launched to the public in November 2022, when it was powered by GPT-3.5, and kicked off the generative AI boom that now dominates Silicon Valley. ChatGPT boasts over 700 million users. Today, powerful large language models are integrated into everything from regular chatbots to agentic assistants and photorealistic video generators. OpenAI's post-training lead, Yann Dubois, said GPT-5 opens up "a whole new world of vibe coding," or the ability for noncoders to create fully deployable websites and apps just by describing what they want to build and letting an AI agent do the work. Altman said he believes GPT-5's ability to "instantaneously create an entire piece of computer software for you" will likely be a core part of its appeal and use. "I think this idea of software on demand will be a defining part of the new GPT-5 era," he said. Earlier this summer, Altman compared contemporary AI agents to an "intern that can work for a couple of hours." He went on to predict that "at some point, it'll be like an experienced software engineer that can work for days." Such tools have already begun to displace white-collar jobs, rousing distress from job seekers. GPT-5 dropped amid an already buzzy week for AI companies. OpenAI had just released two new open-weight models, or models that other developers can download and use for free, on Tuesday. Also on Tuesday, Anthropic dropped Claude Opus 4.1, an upgrade to its powerful Opus 4 model, and Google released its interactive world simulation model Genie 3. Nick Turley, OpenAI's head of ChatGPT, said the company is confident GPT-5 will be more accurate and produce fewer hallucinations, a term for false or misleading information that AI models present as fact. He said the company has made significant improvements on toning down sycophancy, a problem that has plagued ChatGPT in the past. The AI giant recently rolled back an update to GPT-4o that made the bot so overly agreeable that it praised users to the point of sounding disingenuous, and even went so far as to endorse and give instructions for terrorism. Just this week, OpenAI also added guardrails to make its models refrain from giving direct advice about personal challenges. GPT-5 went through 5,000 hours of red teaming, or expert-led safety testing, according to the company's safety research lead, Alex Beutel. Beutel added that the model will be less deceptive than previous versions, meaning it should be less likely to lie to users about things like completing a task it hadn't actually completed. OpenAI also developed a new method, which it dubbed "safe completions," to prevent people from trying to get around safety guardrails by tailoring their questions in ways that trick the model into aiding in a harmful request. "So if someone [asks] how much energy is needed to ignite some specific material, that could be an adversary trying to get around the safety protections and cause harm, or it could be a student asking a science question to understand the physics of this material," Beutel said. "And this creates a real challenge. What's the best way for the model to reply?" In the past, ChatGPT would either respond or refuse to respond based on whether it thinks the question is safe to answer. GPT-5, however, will take a new approach: Rather than choosing between compliance and refusal, it will try to always "give as helpful an answer as possible, but within the constraints of remaining safe." Online, there's been some buzz around what GPT-5 will reveal about humanity's ability to achieve "artificial general intelligence," or AGI, a hypothetical benchmark at which point AI will be fully capable of doing any intellectual task that a human can. Altman, who has been vocal about his optimism for the potential to reach AGI, said GPT-5 isn't there just yet. While the model is generally highly intelligent, he said, it's still missing a key ability to learn continuously without being retrained. "But the level of intelligence here, the level of capability, it feels like a huge improvement," Altman said. "And certainly, if I could go back five years before GPT-3 and tell me we have this now, I'd be like, 'That's a significant fraction of the way to something very AGI-like.'"
[43]
Here's why OpenAI's ChatGPT-5 drop matters
ChatGPT-5 is an OpenAI model that will be able to perform some tasks and make decisions without any human intervention. OpenAI is dropping the long-awaited GPT-5, the latest model of its popular artificial intelligence (AI) chatbot ChatGPT, on Thursday. The new model has advanced writing capabilities, invents answers less often when it doesn't have the right information to a query, and responds better to health-related questions by flagging potential medical concerns, according to the company. GPT-5 has "agentic" capabilities that it will use when developing code, meaning that it will be able to perform some tasks and make decisions without human intervention. The updated model can make websites, apps, and games, and can be integrated with users' Google tools, like Gmail and Google Calendar, OpenAI said. People using a research preview version of ChatGPT will also be able to select between four "personalities" for the chatbot. AI enthusiasts saw that the drop was coming after a since-deleted description for the upgrade on crowdsource coding platform Github was published in the early hours of Thursday morning. The launch of GPT-5 follows OpenAI's introduction of the open-source gpt‑oss series earlier this week. It's the first model where the company has also released the source code since 2019's GPT-2, so it can be fine-tuned by users to meet their specific needs. Why does this launch matter? The launch is the latest development in the road to agentic AI, where agents can take multi-step actions and use tools to browse the web without human prompts. That could look like an AI agent at a customer service centre having the ability to automatically ask questions, look up information in internal documents, respond with a solution, or refer the customer to a human if necessary, according to one example given by Amazon. Earlier this year, OpenAI CEO Sam Altman predicted that the first AI agents will soon be "joining the workforce" and will "materially change the output of companies". In a June blog post, Altman used the example of an AI agent in a software engineering role, saying the agent will "eventually be capable of doing most things a software engineer at a top company with a few years of experience could do, for tasks up to a couple of days long". "It will not have the biggest new ideas, it will require lots of human supervision and direction, and it will be great at some things and bad at others," he wrote. Agents could also be considered "real-but-relatively-junior" virtual coworkers that could be scaled to "every field of knowledge work," he continued. The OpenAI launch of agentic agents would not be the first from an AI company. Microsoft and Google have already launched AI agents that can be customised based on what services a company needs to provide or what tasks it wants to automate.
[44]
GitHub Accidentally Leaks OpenAI GPT-5 Details | AIM
According to the blog, GPT-5 introduces improved reasoning, code generation, and agentic task handling. A blog post on GitHub has revealed the release of OpenAI's latest AI model series, GPT-5, through GitHub Models. The now archived blog post discloses that GPT-5 is going to be available for use across a range of applications, with several variants targeting different use cases. According to the blog, GPT-5 introduces improved reasoning, code generation, and agentic task handling. "It handles complex coding tasks with minimal prompting, provides clear explanations, and introduces enhanced agentic capabilities," the post stated. There are four model variants of GPT-5. The standard gpt-5 model is suited for logic and multi-step reasoning tasks. gpt-5-mini targets cost-sensitive deployments, while gpt-5-nano is optimised for speed and low-latency requirements. For enterprise use ca
[45]
OpenAI's GPT-5 Challenges Claude Opus 4.1, Lags Behind Grok-4 | AIM
"Grok 5 will be out before the end of this year, and it will be crushingly good." After two years of anticipation, OpenAI has finally launched GPT-5, and it's full of surprises. Their most advanced AI model yet is now available to all ChatGPT users, even those on the free plan. The new system introduces a layered reasoning architecture and marks a shift in how users interact with AI for writing, coding, health, and multimodal tasks. "GPT‑5 is a significant leap in intelligence over all our previous models," OpenAI said in the official announcement. The model is now the default option in ChatGPT, replacing GPT-4o and all earlier variants. OpenAI CEO Sam Altman compared GPT-5 to having instant access to a group of PhD-level experts."People are limited by ideas, but not really the ability to execute, in many new ways," he said. One system, m
[46]
OpenAI unveils GPT-5 model, featuring improved coding and problem-solving chops
OpenAI on Thursday unveiled its highly anticipated GPT-5, a powerful multi-modal AI model featuring major advancements in problem-solving and coding. The new flagship model was announced during a Thursday morning livestream. Unlike previous releases that were limited to paid subscribers, GPT-5 will be available to free-tier ChatGPT users as well, OpenAI said. OpenAI CEO Sam Altman described interacting with the new model as akin to conversing with a PhD-level expert, noting that while GPT-3 was comparable to a high school student and GPT-4 to a college student, "With GPT-5 you get an entire team of PhD experts in your pocket, ready to help you," Altman said. During the announcement, OpenAI researchers emphasized that GPT-5 was designed to be more reliable and accurate, with fewer hallucinations.
[47]
OpenAI unveils GPT-5, a new flagship AI model with high accuracy and coding power - SiliconANGLE
OpenAI unveils GPT-5, a new flagship AI model with high accuracy and coding power In a highly-anticipated announcement today, OpenAI released GPT-5, the company's most recent state-of-the-art artificial intelligence model that outperforms previous models on intelligence benchmarks and answers questions which high accuracy. "GPT-5 is a major upgrade over GPT-4o and a significant step along our path to AGI," said Sam Altman, chief executive and co-founder of OpenAI. "GPT-3 was sort of like talking to a high school student. There were flashes of brilliance lots of annoyance but people start to use it and get some value out of it." OpenAI emphasized GPT-5's significant leap in capabilities, particularly in coding, front-end design, and debugging large codebases. The model also delivers a deeper contextual understanding and expressive depth for writing and report generation, areas where earlier models occasionally faltered. Under the hood, GPT-5 is a hybrid system. It routes between a standard model for direct answers and a "thinking" model for deeper reasoning. Depending on the complexity of the user's prompt, GPT-5 automatically decides which model to engage, with an option for users to manually enable the "thinking" mode via model picker or by typing instructions such as "think hard about this." In evaluations, OpenAI said GPT-5 is showed much a significant increase in intelligence over previous models in performance benchmarks, especially in math, coding, visual perception and health. In math performance, the company noted it sets a new state of the art bar with 94.6% on AIME 2025 without tools, for coding 74.9% on SWE-bench Verified and 88% on Aider Polyglot, for multimodal understanding 84.2% on MMMU, and for health 46.2% on HealthBench Hard. The company said these scores appear during everyday use not just in competition. This is nice to see in comparison to the previous models, but how does it hold up against the competition? Anthropic's recent Claude Opus 4.1 model scored 74.5% on SWE-bench Verified, slightly below GPT-5, and Google LLC's Gemini 2.5 Pro scored 59.6%. "GPT-5 as a language model shows continued progress in making AI more useful for real work," Arvind Jain, founder and CEO of Glean Technologies Inc. told SiliconANGLE in an email. "What's notable is that reasoning over data, not just planning, is what drives the model's accuracy." Comparatively, on Humanity's Last Exam -- a benchmark testing general intelligence across disciplines -- a version of GPT-5 with superior reasoning, GPT-5 Pro, scored 42% with tools, just behind xAI Inc.'s Grok 4 Heavy at around 44%. However, users might be more interested in what OpenAI has done about critical issues that affect them more directly such as hallucinations, or the model's likelihood to confabulate and completely make up falsehoods. According to the company, with web search enabled GPT-5 about 45% less likely to generate factual errors than GPT-4o, and when thinking is enabled that is further reduced by about 80% less than o3. The company also said it addressed the "sycophancy problem," an issue that struck GPT-4o earlier this year where it would excessively and overly agree with users (sometimes to their detriment). This included making it less agreeable and use fewer unnecessary emojis -- unless users ask it to, of course. GPT-5 is rolling out today as the new default model for signed-in ChatGPT users, replacing GPT-4o. It auto-switches between reasoning and non-reasoning modes, while paid users can manually enable deeper reasoning. The OpenAI team noted that AI models are beginning to saturate benchmarks and that not everything can be displayed by sheer numbers. To showcase how GPT-5 has become a better "brain" for agentic coding and developing applications, they demonstrated it in a real world situation where it was tasked with debugging a software audio problem. After only a minute of work and thinking on the problem, the model came back with a working, bug-free solution, complete with code changes. "GPT-5 is the smartest coding model we've used. Our team has found GPT-5 to be remarkably intelligent, easy to steer, and even to have a personality we haven't seen in any other model," said Michael Truell, co-founder and chief executive of Anysphere Inc., the maker of Cursor, an AI agent-based coding platform. "It not only catches tricky, deeply-hidden bugs but can also run long, multi-turn background agents to see complex tasks through to the finish." According to OpenAI, GPT-5 has been designed to become a better collaborator, particularly in agentic coding products such as Cursor, Windsurf, GitHub Copilot and CodexCLI. These are platforms where users prompt an AI model with a description of what they want it to do and then let it loose on their codebase, or a blank slate, and allow it to run on its own. AI agents are capable of autonomously performing tasks by breaking down complex workflows to achieve goals by breaking them down into step-by-step plans and then executing them with little or no human intervention. They can also collaborate with a human developer similar to a pair programmer to assist them in producing a product to handle an intricate task. GPT-5 in particular, OpenAI researchers said, has been designed to improve tool calling and follow instructions to swiftly perform coding tasks according to natural language prompts. This makes it ideal for what is known as "vibe coding," an emerging software development trend where developers use AI to assist them by generating code from prompts, rather than writing it manually. With the recent changes to the model and the additional capabilities have made the model better at reasoning out creative performance. This includes understanding color, user interface design and user intent. "GPT-5 really brings the power of beautiful and effective code to everyone." said Yan Dubois, OpenAI researcher. For developers, GPT-5 comes in three tiers -- GPT-5, GPT-5-mini and GPT-5-nano -- offered via application programming interface with options for cost, latency and reasoning depth. Tool call preambles, verbosity controls and regex-enforced outputs are now part of the package, making it even more reliable and tunable than previous models. "For me it's the first time I trust a model to do my most important work," said Yan Dubois, solutions architect at OpenAI. "This is beyond vibe coding."
[48]
OpenAI releases long-awaited GPT-5 AI chatbot upgrade
GPT-5, the long-awaited upgrade to the ChatGPT AI chatbot, has been released by its maker OpenAI. It has been one of the most highly anticipated launches in Silicon Valley after OpenAI's first offering ChatGPT - powered by its GPT-3 model - kick-started the current AI boom in late 2022. "GPT-3 sort of felt like talking to a high school student," said Sam Altman, OpenAI's chief executive. "GPT-4, maybe it was like talking to a college student. But with GPT-5, now it's like talking to an expert, a PhD-level expert in anything, any area you need, on demand." At the launch event, OpenAI claimed the new chatbot, which will be released to all ChatGPT users on Thursday, was more than a simple upgrade to its previous offerings. According to OpenAI, the new model exceeds the chatbot competition from the likes of Google, X and Antropic on "benchmarks" - standardised tests used to rank models. OpenAI claims it has been designed to be easier and more natural to communicate with, better at writing prose and advanced computer code, solving academic questions from mathematics to law, assisting with healthcare-related questions, as well as being safer than its predecessors. "It's an incredible superpower on demand," claimed Mr Altman. The model is also more intelligent in how it uses its own brain power - and therefore an expensive computing resource - according to OpenAI. It is a hybrid of previous chatbots and slower, more computing-intensive "reasoning" models like OpenAI's Deep Research. Based on a user's request, the model will decide how much "thinking" is required before answering, rather than requiring the user to switch between different models. Although AI enthusiasts who had been expecting GPT-5 to represent "artificial general intelligence [AGI]" will be disappointed. Despite this being OpenAI's stated goal, Mr Altman billed GPT-5 as a "major upgrade" to GPT-4 and a "significant step along the path to AGI". OpenAI is projected to spend $8bn (£6bn) this year, on top of $5bn (£3.7bn) last year, and while it is expected to make a profit this year, the business case for increasingly powerful AI models is still not clear to many investors. Given a single training run for GPT-5 is rumoured to have cost $500m (£373m), there will be an expectation the new model is significantly more useful to business users. Despite a very slick demonstration of its coding skills at the launch presentation, where it built an online language learning game in seconds, GPT-5 will have to prove its worth for professional coding. Many in the tech industry prefer Anthropic AI's Claude model to write code. OpenAI and its investors will be hoping GPT-5 changes that. AI experts will also be testing GPT-5's tendency to "hallucinate", an issue OpenAI claims to have improved with GPT-5. But erroneous or bizarre answers are a problem that dogs all large generative AI models. "Shiny things are always fun to play with, and I fully expect GPT-5 to be the shiniest so far," said Gary Marcus, a cognitive scientist at New York University and AI commentator. "But that doesn't mean that it is a critical step on the optimal path to AI that we can trust."
[49]
GPT-5 debuts with new ChatGPT personalities to curb sycophancy
OpenAI announced its foundational large-language model, GPT-5, revealing four new personality options for ChatGPT to reduce sycophancy, rolling out today, despite an ongoing lawsuit from Ziff Davis. On Thursday, OpenAI introduced GPT-5, a new foundational large-language model. This model integrates numerous ChatGPT features and reasoning capabilities into a unified system. This integration aims to facilitate deep research, the creation of custom applications, and the completion of various tasks for users. As part of this announcement, OpenAI also disclosed the introduction of four new personality options for its AI chatbot, ChatGPT. These personalities, named Cynic, Robot, Listener, and Nerd, will initially be accessible for text-based interactions and will subsequently be extended to voice chat. OpenAI stated in a blog post that these new preset personalities are a direct result of improvements in steerability. The company specified, "We're also launching a research preview of four new preset personalities for all ChatGPT users, made possible by the improvements on steerability. These personalities, available initially for text chat and coming later to Voice, let you set how ChatGPT interacts -- whether concise and professional, thoughtful and supportive, or a bit sarcastic -- without writing custom prompts." These new personality features are opt-in, allowing users to choose whether to activate them. Users also have the ability to adjust these personalities, enabling fine-tuning of the chatbot's interactive style to align with specific requirements or preferences. A primary objective behind introducing these new personalities is to mitigate the sycophancy that has been observed in recent iterations of ChatGPT. Previous versions of ChatGPT developed a reputation for being excessively agreeable, which led to user experience issues. The New York Times reported instances where some users found that ChatGPT's overly compliant nature encouraged delusions and potentially exacerbated mental health crises. This issue, combined with the documented high hallucination rates in some ChatGPT models, contributed to a suboptimal user experience. OpenAI's blog post addressed this concern directly, stating, "All of these new personalities meet or exceed our bar on internal evals for reducing sycophancy." The post further elaborated on the general behavior of the new model, noting, "Overall, GPT‑5 is less effusively agreeable, uses fewer unnecessary emojis, and is more subtle and thoughtful in follow‑ups compared to GPT‑4o." The company aims for the interaction with GPT-5 to feel "less like 'talking to AI' and more like chatting with a helpful friend with PhD‑level intelligence." GPT-5 is scheduled for a rollout today, making it accessible to users shortly after the announcement. It is important to note that Ziff Davis, the parent company of Mashable, filed a lawsuit against OpenAI in April. The lawsuit alleges that OpenAI infringed Ziff Davis copyrights during the training and operation of its AI systems.
[50]
OpenAI releases GPT-5, the latest model fueling ChatGPT
ChatGPT is set to reach 700 million weekly active users, OpenAI previously said. OpenAI on Thursday released GPT-5, unveiling a new version of the artificial intelligence model that fuels popular chatbot ChatGPT. The announcement comes days after the company said ChatGPT is set to reach 700 million weekly active users, up from 500 million just five months ago. OpenAI CEO Sam Altman described GPT-5 as a "legitimate PhD expert in any area," saying previous versions of the AI model amounted to college or high school students by comparison. GPT-5 enables users to write computer programs, prepare party invitations and better navigate the health care system, Altman said. "It is useful, it's smart, it's fast, it's intuitive," Altman said. "It's an incredible superpower on demand, unimaginable at any other time in history."
[51]
GPT-5 is officially out
OpenAI unveiled its new frontier model, GPT-5, marking a significant advancement in artificial intelligence, making it available to ChatGPT users and via API, following weeks of anticipation. The latest iteration of the technology that powers ChatGPT is designed to be faster, smarter, and more useful than its predecessors. The company reports that GPT-5 exhibits lower hallucination rates and features enhanced agentic capabilities. This model will be accessible to all ChatGPT users, including those on the free-tier, albeit with specified usage limits. Additionally, it will be available through the API, allowing other application developers and businesses to integrate and build upon its functionalities. GPT-5 demonstrates improved performance in practical computer programming and mathematical tasks compared to OpenAI's existing models. The model can also execute multi-step tasks independently, recover from errors more effectively than previous versions, and efficiently call external tools. Its capacity to persist through complex workflows represents a significant leap, which developers are finding particularly beneficial in the creation of real-world applications. Sam Altman, CEO of OpenAI, described the improvement during a press briefing, stating, "GPT-3 felt like talking to a high school student; GPT-4 felt like a college student, but GPT-5 is the first time it really feels like talking to an expert." He further elaborated on the perceived advancement, adding, "It reminds me of when the iPhone went from those giant pixels to [the] Retina display... I can't believe how bad we had it." The release of GPT-5 follows a period of notable activity for OpenAI. Earlier in the week, specifically on Tuesday, the company introduced two free open-weight models capable of running on edge devices. On Wednesday, OpenAI announced a new initiative to offer ChatGPT to government entities for $1 per agency. Furthermore, the company has indicated plans to introduce new mental health-related features designed to encourage healthier digital habits and assist in detecting user mental or emotional distress. OpenAI initially introduced generative AI to the public with ChatGPT approximately three years ago, which rapidly became one of the fastest-growing applications. This was followed by GPT-3.5 and later GPT-4, with GPT-4 demonstrating substantial progress in intelligence and surpassing human performance in various tasks. ChatGPT is nearing 700 million weekly active users as of this week, an increase from 500 million users reported in March, indicating an over fourfold surge in growth year-over-year. Concurrently, OpenAI recently secured $8.3 billion in a new round of venture capital funding, establishing a valuation of $300 billion. The company is also exploring a potential secondary stock sale, which could potentially value it at $500 billion. The deployment of GPT-5 encountered several delays, attributed to factors such as more rigorous safety checks and challenges related to server capacity. Model developers, including OpenAI, have faced difficulties in demonstrating major performance gains from subsequent AI models like GPT-4.5 in recent months. Researchers have observed a deceleration in the pace of AI advancement, resulting in each new model offering only marginal improvements over its predecessor. GPT-5 addresses some of these concerns, according to Altman. Regarding safety, OpenAI has stated that GPT-5 is considerably less prone to hallucinating or fabricating information. The model now employs a new system called "safe completions" for handling sensitive inquiries, providing helpful responses within established safety parameters and offering clear explanations when it cannot provide assistance. The new model also integrates OpenAI's flagship "GPT"-branded large language models with its "o" line of reasoning models into a single model or chat interface. This integration follows Altman's earlier observation that the company's product offerings had become "complicated." GPT-5 also signifies OpenAI's increased investment in "test-time compute." This approach involves channeling more processing power to enable models to solve more challenging tasks, an area Altman has indicated OpenAI would prioritize. Altman commented on this investment, stating, "We are discovering new paradigms that I think will be as powerful as scaling laws in the past -- this idea that we can use more compute, higher quality data and better environments to make smarter models, we see orders of magnitude more gains in front of us." He further acknowledged the financial commitment required, adding, "Obviously, we have to invest in compute at an eye-watering rate to get that, but we intend to keep doing it." Ultimately, Altman stated that GPT-5 represents a step closer to OpenAI's objective of achieving Artificial General Intelligence (AGI), though he noted that comprehensive AGI has not yet been realized. He elaborated, "This is clearly a model that is generally intelligent, although I think the way that most of us define AGI, we're still missing many things [that are] quite important." A key missing element, he specified, is continuous learning, stating, "But one big one is, this is not a model that continuously learns as it's deployed for new things it finds, which is something that, to me, feels like it should be part of an AGI."
[52]
OpenAI unveils GPT-5
OpenAI on Thursday unveiled its new version of ChatGPT, GPT-5, with the company's CEO Sam Altman calling it "a major upgrade" from the previous version. "Thirty-two months ago, we launched ChatGPT. And since then, it has become the default way that people use AI. In that first week, a million people tried it out, and we thought that was pretty incredible," Altman said in a Thursday livestream. "But now, about 700 million people use ChatGPT every week, and increasingly rely on it to work, to learn, for advice, to create and much more. Today, finally, we're launching GPT-5. GPT-5 is a major upgrade over GPT-4," he continued. According to OpenAI, the new version of ChatGPT will see advancement in areas such as coding, writing and health. For example, the company claims the new version of the tool "more reliably handles writing that involves structural ambiguity, such as sustaining unrhymed iambic pentameter or free verse that flows naturally." "GPT‑5 not only outperforms previous models on benchmarks and answers questions more quickly, but -- most importantly -- is more useful for real-world queries," the company states on its website. "We've made significant advances in reducing hallucinations, improving instruction following, and minimizing sycophancy, while leveling up GPT‑5's performance in three of ChatGPT's most common uses: writing, coding, and health." In the last few years, the artificial intelligence race has consumed the tech industry, with big names like Google and Microsoft throwing their hats into the ring for the technology alongisde Wall Street investment. On Tuesday, OpenAI said it will offer ChatGPT to federal agencies for $1 every year in partnership with the General Services Administration (GSA). The announcement followed the agency adding OpenAI's artificial intelligence model to its government purchasing system the day before.
[53]
OpenAI Releases GPT-5, a Potential Barometer for Whether Artificial Intelligence Hype Is Justified
SAN FRANCISCO (AP) -- OpenAI has released the fifth generation of the artificial intelligence technology that powers ChatGPT, a product update that's being closely watched as a measure of whether generative AI is advancing rapidly or hitting a plateau. GPT-5 arrives more than two years after the March 2023 release of GPT-4, bookending a period of intense commercial investment, hype and worry over AI's capabilities. In anticipation, rival Anthropic released the latest version of its own chatbot, Claude, earlier in the week. Expectations are high for the newest version of OpenAI's flagship model because the San Francisco company has long positioned its technical advancements as a path toward artificial general intelligence, or AGI, a technology that is supposed to surpass humans at economically valuable work. It is also trying to raise huge amounts of money to get there, in part to pay for the costly computer chips. OpenAI started in 2015 as a nonprofit research laboratory to safely build AGI and has since incorporated a for-profit company with a valuation that has grown to $300 billion. The company has tried to change its structure since the nonprofit board ousted its CEO Sam Altman in Nov. 2023. He was reinstated days later and continues to lead OpenAI. It has run into hurdles escaping its nonprofit roots, including scrutiny from the attorneys general in California and Delaware, who have oversight of nonprofits, and a lawsuit by Elon Musk, an early donor to and founder of OpenAI. Most recently, OpenAI has said it will turn its for-profit company into a public benefit corporation, which must balance the interests of shareholders and its mission.
[54]
OpenAI Just Released a New ChatGPT That's 'Much Smarter Across the Board,' According to Its CEO. Here Are Some GPT-5 Prompts to Get You Started.
GPT‑5 "is less effusively agreeable" and "uses fewer unnecessary emojis," the company says. OpenAI unveiled GPT-5 on Thursday, which the company is calling its "smartest, fastest, [and] most useful model yet." In a press release, the company said the new model puts "expert-level intelligence in everyone's hands" and is a "significant leap in intelligence over all our previous models." On Wednesday, OpenAI CEO Sam Altman called it a "major upgrade" and said that after the new model, using older versions felt "miserable." Related: OpenAI CEO Sam Altman Says AI Agents Are Like a Team of 'Junior Employees' In addition to improvements in coding, math, health, and other areas, the new GPT-5 also lets users choose a "personality" in settings. The default personality is still available and noted to be "clear, neutral, and adaptable." Other options include "cynic" (sarcastic and dry, blunt but witty); "robot" (precise, emotionless, direct answers without extra words); "listener" (warm, laid-back, and calm with "light wit"); and "nerd" (playful, curious, celebrates knowledge and discovery). And while previous reports highlighted the sycophancy of past models, OpenAI says that GPT‑5 "is less effusively agreeable" and "uses fewer unnecessary emojis." "It should feel less like 'talking to AI' and more like chatting with a helpful friend with PhD‑level intelligence," the company wrote. Related: Here's What 'Terrifies' OpenAI's CEO About Financial Institutions Today GPT-5 is available for all users. OpenAI recommends these prompts to get started:
[55]
GitHub May Have Accidentally Leaked OpenAI's Upcoming GPT-5 Models
Evidence for GPT-5 Pro and Thinking variants have also been spotted OpenAI's upcoming artificial intelligence (AI) model, GPT-5, has reportedly been leaked ahead of release. As per the report, the models were listed in a now-deleted GitHub blog post, along with certain details about them. The post reportedly mentions four variants, which include the previously reported GPT-5 base model, GPT-5 Mini, GPT-5 Nano, and a new GPT-5 Chat, which is said to be for enterprise usage. Additionally, a GPT-5 Pro model will reportedly also be released for the company's Pro subscribers. Alongside, ChatGPT is also said to get a redesigned interface. TestingCatalog spotted a GitHub blogpost that accidentally leaked several details about the upcoming GPT-5 models. The post appears to have been taken down, but not before users captured and shared screenshots on social media platforms. Based on that, it appears that OpenAI is planning to release four variants of its next-generation frontier model. The base variant of OpenAI's GPT-5 model will reportedly be accompanied by a Mini and a Nano variant. While no details about them are known, it is likely that Mini offers lower latency response generation, and Nano could be designed for edge and local deployments. Apart from this, references have reportedly also appeared for GPT-5 Thinking and GPT-5 Chat. While no details are currently known about the former, this could be an application programming interface (API) offering, allowing developers to adjust the thinking budget for the model. GPT-5 Chat, as per the deleted GitHub post, is designed for GitHub Copilot and will be more context-aware in conversations. The full description of the variant, according to The Verge, reads, "Designed for advanced, natural, multimodal, and context-aware conversations for enterprise applications." As per the report, access to the model will be rolled out to all users, with varying rate limits and restrictions. Those on ChatGPT's free tier will reportedly get basic access to GPT-5, while Plus subscribers are said to get GPT-5 with "advanced reasoning" and higher limits. Pro subscribers will reportedly get a new variant dubbed GPT-5 Pro, as well as even higher rate limits. ChatGPT Team subscribers are said to get access to both GPT-5 with reasoning as well as the GPT-5 Pro model. A separate leak shared by user @scaling01 on X (formerly Twitter) reveals that GPT-5 Pro could offer "research-grade intelligence". The tipster found this description within the codebase of ChatGPT. The functionality of the feature in real-life scenarios remains unclear. TestingCatalog also claims that ChatGPT will get a redesign with the release of GPT-5. The new user interface is said to offer a cleaner prompt bar, a consolidated button housing different actions and modes, and new background colours. In a post on X, OpenAI CEO Sam Altman announced a new livestream on Thursday, which will begin at 10 am PDT (10:30pm IST). While Altman did not reveal what would be announced during the livestream, he did highlight that it would be an hour-long presentation, longer than typical OpenAI showcases. Many expect the company to release GPT-5 during the show.
[56]
OpenAI's livestream today: Will Sam Altman be launching GPT-5? When and where to watch
All signs point to the company gearing up to launch its next-generation AI model, widely anticipated to be GPT-5. The move follows weeks of cryptic hints, subtle remarks, and coded teasers from senior OpenAI executives. In a development that has stirred excitement across the tech world, artificial intelligence powerhouse OpenAI is set to make a major announcement later today. All indicators suggest that the company is preparing to unveil its next-generation AI model, widely expected to be GPT-5. This comes after weeks of hints, veiled statements, and coded teasers shared by top OpenAI executives. The company announced an event scheduled for Thursday, stylized unusually as a "LIVE5TREAM," cleverly substituting the "S" with a "5" -- a subtle yet deliberate nod to the expected GPT-5 release. The OpenAI livestream is set to begin at 10am Pacific Time (10:30pm IST), and will be hosted on the company's official YouTube channel. The speculation around GPT-5 was further fueled when OpenAI CEO Sam Altman recently shared a screenshot revealing "ChatGPT 5" prominently displayed in the top-left corner of the application interface. While the company refrained from confirming the nature of the event outright, the online chatter has become difficult to ignore. Adding to the intrigue, OpenAI's head of applied research also chimed in earlier this week, expressing eagerness about how the public will receive the new model. Even last month, Altman had openly hinted that GPT-5 would be released "soon," although no specific timeline had been committed to, until now. With mounting signals and public statements from OpenAI leadership, industry watchers are almost certain that the highly anticipated GPT-5 will take center stage during today's event. The fifth-generation language model is projected to bring refined capabilities and improved integration features, though insiders are tempering expectations about the leap in performance. Unlike the substantial jump between GPT-3 and GPT-4, early testers report that GPT-5 builds on its predecessor more incrementally. A defining feature of GPT-5, according to sources familiar with early iterations, is its unified reasoning architecture. This upgrade eliminates the need for users to manually select a reasoning model -- GPT-5 is expected to automatically determine and deploy the most appropriate reasoning path based on the user's input. This not only enhances user convenience but potentially sets a new standard in natural language interaction. Some testers noted marked improvement in coding and complex problem-solving abilities, although the broader evolution remains more evolutionary than revolutionary. The run-up to the GPT-5 announcement has been carefully orchestrated. Just two days before the livestream, OpenAI released two new open-weight AI models, gpt-oss-120b and gpt-oss-20b. While these models are significant in their own right, especially with the latter being optimized for local use on consumer-grade PCs, they now appear to be precursors to a larger rollout strategy. The staged releases may be part of a broader campaign hinted at by Altman himself, who recently said the company is gearing up to launch a "ton of stuff" over the coming months. This includes not just new models, but products, tools, and feature upgrades across OpenAI's ecosystem. It is anticipated that today's GPT-5 event will be just the first in a series of major announcements from the company, setting the tone for what could be an eventful close to 2025. OpenAI is sticking to a familiar template with its livestream, using both its YouTube channel and potentially its X page (formerly Twitter) to reach audiences. This approach mirrors previous rollouts such as those for ChatGPT-4 and other key model releases. The event will commence at 1 PM Eastern Time (10 AM PT / 10:30 PM IST), and viewers from around the globe are expected to tune in. Given the global interest in generative AI and the widespread application of previous GPT models, the unveiling could draw one of the largest live audiences for a tech event this year. Despite the fanfare, Altman has already struck a note of caution. In a recent post, he warned that due to potentially high demand, users might experience temporary issues accessing the new model immediately after launch. This is not uncommon for OpenAI, which has seen servers strained during earlier rollouts. Such capacity constraints may delay general availability or cause intermittent access as the company scales up its infrastructure to accommodate the flood of interest. Nonetheless, expectations remain high, particularly for users in sectors like education, software development, research, journalism, and customer service, where GPT models have already become indispensable. While the leap between GPT-4 and GPT-5 may not be as pronounced as the previous generational shift, the introduction of unified reasoning, more seamless user experience, and anticipated performance enhancements still mark a significant evolution in AI development. If early insights prove accurate, GPT-5 could streamline complex workflows, boost coding precision, and improve the relevance of AI responses across disciplines. With the OpenAI livestream just hours away, the world watches keenly to see whether GPT-5 will live up to the hype. For now, all eyes remain on that cleverly stylized "LIVE5TREAM." The OpenAI livestream, teasing what is widely believed to be the launch of GPT-5, will be broadcast at 10 AM Pacific Time / 1 PM Eastern Time / 10:30 PM IST. Viewers can watch it live on OpenAI's official YouTube channel, and possibly via the company's page on X. OpenAI is expected to unveil GPT-5, its next-generation AI model, during a livestream event called "LIVE5TREAM." The stylized name -- with the "5" in place of the "S" -- has fueled speculation that GPT-5 is the centerpiece of the announcement. Yes, but with caution. CEO Sam Altman has warned about possible temporary access issues due to high demand, similar to previous rollouts. Users may experience delays or limited access until infrastructure scales up.
[57]
OpenAI ChatGPT-5 release by Sam Altman: Features, how is GPT-5 different from GPT-3, GPT-4?
Nearly three years ago, ChatGPT introduced the world to generative AI, dazzling users with its ability to write humanlike prose and poetry, quickly becoming one of the fastest growing apps ever. OpenAI ChatGPT-5 is se to be launched on Thursday. The latest installment of the AI technology that powered the ChatGPT juggernaut in 2022, is set for an imminent release, and users will scrutinize if the step up from GPT-4 is on par with the research lab's previous improvements. Two early testers of the new model told Reuters they have been impressed with its ability to code and solve science and math problems, but they believe the leap from GPT-4 to GPT-5 is not as large as the one from GPT-3 to GPT-4. The testers, who have signed non-disclosure agreements, declined to be named for this story. GPT-4's leap was based on more compute power and data, and the company was hoping that "scaling up" in a similar way would consistently lead to improved AI models. But OpenAI, which is backed by Microsoft and is currently valued at $300 billion, ran into issues scaling up. One problem was the data wall the company ran into, and OpenAI's former chief scientist Ilya Sutskever said last year that while processing power was growing, the amount of data was not. He was referring to the fact that large language models are trained on massive datasets that scrape the entire internet, and AI labs have no other options for large troves of human-generated textual data. Apart from the lack of data, another problem was that 'training runs' for large models are more likely to have hardware-induced failures given how complicated the system is, and researchers may not know the eventual performance of the models until the end of the run, which can take months. "OpenAI made such a great leap from GPT-3 to GPT-4, that ever since then, there has been an enormous amount of anticipation over GPT-5," said Navin Chaddha, managing partner at venture capital fund Mayfield, who invests in AI companies but is not an OpenAI investor. "The hope is that GPT-5 will unlock AI applications that move beyond chat into fully autonomous task execution." Nearly three years ago, ChatGPT introduced the world to generative AI, dazzling users with its ability to write humanlike prose and poetry, quickly becoming one of the fastest growing apps ever. In March 2023, OpenAI followed up ChatGPT with the release of GPT-4, a large language model that made huge leaps forward in intelligence. While GPT-3.5, an earlier version of the model, received a bar exam score in the bottom 10 per cent, GPT-4 passed the simulated bar exam in the top 10 per cent. GPT-4 then became the model to beat and the world came to terms with the fact that AI models could outperform humans in many tasks. Soon, other companies were catching on. The same year, Alphabet's Google and Anthropic - which is backed by Amazon and Google - released competitive models to GPT-4. Within a year, open-source models on par with GPT-4 such as Meta Platforms' Llama 3 models were released. Along with training large models, OpenAI has now invested in another route, called "test-time compute," which channels more processing power to solve challenging tasks such as math or complex operations that demand human-like reasoning and decision-making. The company's CEO Sam Altman said earlier this year that GPT-5 would combine both test-time compute and its large models. He also said that OpenAI's model and product offerings had become 'complicated'. Q1. Who is OpenAI CEO? A1. OpenAI CEO is Sam Altman. Q2. What is OpenAI ChatGpt? A2. Nearly three years ago, ChatGPT introduced the world to generative AI, dazzling users with its ability to write humanlike prose and poetry, quickly becoming one of the fastest growing apps ever.
[58]
OpenAI's next-generation ChatGPT-5 is here -- is this the AI moment we've all been waiting for?
OpenAI GPT-5 is finally here, and it's the most powerful AI model OpenAI has ever released. With smarter memory, deeper reasoning, and more human-like conversations, GPT-5 is already changing how we use AI at work, in schools, and in creative projects. This next-gen model boosts ChatGPT's performance and brings advanced safety and better personalization to the platform. From writing and coding to images and videos, GPT-5 takes everything to the next level. Whether you're a business owner, student, or tech lover, there's something big waiting in GPT-5 that you don't want to miss.
[59]
ChatGPT-5 launched: A modular, multi-modal leap -- will OpenAI's most powerful AI transform coding, health, and education forever now?
ChatGPT-5 launched and it's already creating waves across tech, health, and education. OpenAI's latest and most advanced AI model is a modular, multi-modal breakthrough, capable of handling complex tasks with unmatched speed and context. From helping developers write better code to assisting doctors in diagnostics and transforming online learning -- ChatGPT-5 is set to redefine the way we interact with machines. Its new capabilities go far beyond text, opening a new era of smart tools and digital intelligence.
[60]
OpenAI introduces ChatGPT 5 - Here's all you need to know
OpenAI has officially launched GPT-5, marking a major milestone in the evolution of artificial intelligence. Representing a significant leap in capability and intelligence, GPT-5 sets a new standard with its state-of-the-art performance across a wide range of domains, including coding, mathematics, writing, healthcare, visual understanding, and more. What sets GPT-5 apart is its ability to adapt its thinking in real time it can respond instantly when needed, but also knows when to pause and reason deeply, delivering expert-level insights with precision. This makes it the most refined and reliable AI model OpenAI has released to date. GPT-5 is now available to all users. Plus subscribers benefit from increased usage limits, while Pro subscribers gain access to GPT-5 Pro a specialized version of the model offering enhanced reasoning capabilities for more comprehensive, accurate, and context-aware responses. GPT-5 represents a significant leap forward in both performance and intelligence. Built as a unified system, it adapts seamlessly to user needs: offering fast responses for simple questions, switching to deeper reasoning when faced with complex problems, and intelligently choosing between the two using a real-time decision router. This router trained continuously using real-world feedback like user preferences, model switches, and accuracy ratings ensures that every response is as helpful and relevant as possible. Even when usage limits are reached, lighter versions of the model step in to maintain functionality without sacrificing quality. Future updates will further streamline these capabilities into a single, integrated model. GPT-5 offers state-of-the-art results across key areas: Writing: More coherent, personalized, and stylistically aware. Math: Higher accuracy with step-by-step reasoning for complex problems. Health: Improved safety, contextual understanding, and usefulness for health-related queries. Visual Perception: Stronger image interpretation and description. Instruction Following: Better at understanding nuanced prompts, delivering more useful and grounded results. Crucially, GPT-5 also addresses persistent challenges seen in earlier models. It reduces hallucinations, improves factual consistency, follows instructions more closely, and limits sycophantic or overly agreeable responses -- making it a more trustworthy assistant in high-stakes scenarios. With GPT-5, OpenAI introduces its strongest coding model yet. It excels at front-end generation, debugging large codebases, and design-focused development. Whether building responsive websites or intuitive apps, GPT-5 often delivers high-quality results in a single prompt complete with clean layouts, elegant typography, and smart spacing. Developers using GPT-5 have reported major improvements in its ability to handle both logic and design fluently, significantly reducing iteration time and improving output quality. Full developer tools and documentation are available for those looking to unlock its capabilities in technical workflows. GPT-5 is available to all ChatGPT users. Plus subscribers enjoy expanded access with higher usage limits. Pro subscribers get GPT-5 Pro a premium version with extended reasoning abilities, designed for deeper analysis and even more accurate results. As OpenAI continues to refine GPT-5's thinking architecture, the company remains committed to ensuring AI is both powerful and safe, useful not just in theory but in everyday life, across industries and disciplines. OpenAI has positioned GPT-5 with aggressively competitive pricing, aiming to make its most advanced capabilities accessible to a wider range of users and businesses. Here's how the pricing breaks down: GPT-5: $1.25 per million input tokens $10.00 per million output tokens GPT-5 Mini: $0.25 per million input tokens $2.00 per million output tokens GPT-5 Nano: $0.05 per million input tokens $0.40 per million output tokens Compared to GPT-4o, GPT-5 cuts input costs in half, while maintaining the same output rate. However, users should note that GPT-5's advanced reasoning capabilities involve invisible reasoning tokens, which are billed as output. This means that, unless the reasoning level is explicitly set to "minimal," many prompts may generate higher output token usage than their GPT-4o equivalents. This pricing strategy not only delivers more value per dollar but also reflects GPT-5's smarter use of resources -- prioritizing depth, accuracy, and responsiveness.
[61]
OpenAI drops GPT‑5: The most powerful, human-like AI yet -- are PhD-level minds now just a tap away? Here's pricing and availability
OpenAI GPT-5 is here, and it's a total game-changer in the world of AI. Designed to feel like a team of PhD-level experts in your pocket, GPT-5 brings faster answers, smarter conversations, and powerful features like vibecoding -- which lets you build apps with simple prompts. It's more accurate, more personal, and deeply integrated with tools like Gmail and Calendar. Whether you're writing, coding, or planning your day, GPT-5 makes it easier and faster than ever. With full access for ChatGPT Pro users and limited use for free users, GPT-5 is leading the future of AI -- and it's here to help everyone.
[62]
ChatGPT 5 Incoming : 10 New Expected Features of OpenAI's GPT-5
What if the next leap in artificial intelligence could not only solve your toughest problems but also anticipate your needs before you even voice them? With the much-anticipated release of ChatGPT-5 in a few hours, this vision is closer than ever. OpenAI's latest model is set to redefine what's possible, boasting features like superior reasoning capabilities that outperform human benchmarks, expanded multimodal functionality, and tools to autonomously manage complex workflows. Whether you're a developer looking to streamline coding tasks or a professional seeking reliable, context-aware insights, GPT-5 promises to be more than just an upgrade -- it's a fantastic option. The AI Grid speculate on 10 new features that make GPT-5 a monumental step forward in AI technology. From minimizing hallucinations to processing vast amounts of data with its expanded context window, these advancements aim to address the limitations of previous models while unlocking new possibilities for innovation. But what does this mean for you? How might these features reshape the way you work, create, or even think? Let's unpack the fantastic potential of GPT-5 and consider how it's poised to redefine the boundaries of human-AI collaboration. GPT-5 is expected to significantly improve its coding capabilities, making it a more powerful tool for software development. With a deeper understanding of programming languages, frameworks, and algorithms, it will enable faster and more precise application development. For instance, GPT-5 could assist in debugging complex code, optimizing algorithms, or even generating entire software modules. This advancement is likely to streamline workflows for developers, reduce operational costs, and accelerate time-to-market for software products. Whether you're working on SaaS platforms or mobile applications, these improvements promise to enhance productivity and efficiency. To address inconsistencies in response accuracy, GPT-5 will introduce a "universal verifier" mechanism. This feature is designed to evaluate and refine its outputs, making sure that responses are both accurate and contextually relevant. This improvement is particularly valuable for applications requiring high reliability, such as academic research, professional consultations, or technical support. By delivering precise and well-contextualized answers, GPT-5 aims to become a trusted resource for users seeking dependable information. GPT-5 is set to achieve new advancements in reasoning capabilities, surpassing human performance on benchmarks like Simple Bench, which evaluates spatial, social, and linguistic reasoning. This enhancement will make it an indispensable tool for tackling complex problems in fields such as data analysis, strategic planning, and education. For example, you could rely on GPT-5 to analyze intricate datasets, simulate scenarios requiring nuanced reasoning, or develop strategies for business growth. These capabilities promise to save time, improve decision-making, and expand the scope of AI-assisted problem-solving. One of the most impactful updates in GPT-5 is its expanded context window, reportedly capable of processing up to 1 million tokens. This feature will allow the model to maintain continuity over extended conversations or analyze large datasets, such as legal documents, research papers, or historical archives. For you, this means seamless interactions and the ability to tackle projects requiring long-term memory and contextual understanding. Whether you're drafting a comprehensive report or conducting in-depth research, GPT-5's expanded context window will enhance its utility and effectiveness. GPT-5's enhanced multimodal capabilities will enable it to process and analyze not only text but also audio and video inputs. This opens up new possibilities for applications in media, communication, and accessibility. For instance, you could use GPT-5 to transcribe and summarize video content, generate insights from audio recordings, or even create multimedia presentations. These features make GPT-5 a versatile tool for content creators, educators, and professionals, offering innovative ways to manage and interpret diverse forms of data. Discover other guides from our vast content that could be of interest on ChatGPT-5 features. Efforts to minimize hallucinations -- instances where the model generates incorrect or fabricated information -- are a key focus in GPT-5. This improvement is particularly critical for high-stakes fields like healthcare, law, and finance, where accuracy is paramount. By reducing the likelihood of errors, GPT-5 aims to build greater trust and reliability in its outputs. Whether you're drafting a legal document, analyzing financial data, or seeking medical advice, you can depend on GPT-5 for accurate and trustworthy information. GPT-5 introduces agentic functionality, allowing it to autonomously manage multi-step tasks and proactively call external tools. This means the model can act as a virtual assistant, handling complex workflows without requiring constant supervision. For example, you could assign GPT-5 a project involving data collection, analysis, and reporting, and it would execute these steps independently. This capability has the potential to save you significant time and effort, making GPT-5 an invaluable tool for project management and task automation. Dynamic reasoning capabilities in GPT-5 will allow the model to adjust its computational intensity based on the complexity of the task. This ensures optimal performance by balancing speed and accuracy. For straightforward queries, GPT-5 will provide quick responses, while for more intricate problems, it will allocate additional resources to deliver thorough and well-analyzed answers. This adaptability enhances overall efficiency and ensures that the model meets your specific needs, regardless of the task's complexity. Safety remains a top priority in GPT-5, with advanced safeguards designed to prevent misuse. These include measures to avoid jailbreaking, generating harmful content, or providing unsolicited advice. Additionally, ongoing research into controlling superintelligent AI through weaker models underscores a commitment to responsible AI development. These safety enhancements aim to ensure that GPT-5 operates within ethical boundaries, giving you confidence in its secure and responsible deployment. To cater to diverse needs, GPT-5 will be available in multiple variants, each tailored to specific use cases: These variants ensure that you can select the version best suited to your requirements, whether you prioritize high performance, affordability, or portability. OpenAI ChatGPT-5 represents a significant advancement in artificial intelligence, offering features that enhance reasoning, coding, multimodal capabilities, and safety mechanisms. By addressing existing limitations and introducing innovative functionalities, GPT-5 is poised to become an indispensable tool across industries. Whether you're a developer, researcher, educator, or business professional, these advancements promise to make your interactions with AI more efficient, reliable, and impactful.
[63]
OpenAI Launches GPT-5, Makes It Free for All ChatGPT Users
GPT-5 delivers strong performance across coding, math, visual perception, writing, health, and more. It also dramatically reduces the hallucination. The wait is over! After over two years of waiting, GPT-5 is finally official. OpenAI has launched the GPT-5 AI model, the successor to GPT-4/4o, and it's already rolling out to all ChatGPT users. GPT-5 is a hybrid reasoning model, meaning it can generate both quick answers and reason over complex queries. OpenAI says GPT-5 delivers "state-of-the-art performance across coding, math, writing, health, visual perception, and more." GPT-5 is more like a unified AI system rather than a standalone AI model. The company notes that GPT-5 is the strongest model to date. It can debug larger repositories and create beautiful front-end generations. On top of that, GPT-5 is OpenAI's most capable AI model for writing that can generate "compelling, resonant writing with literary depth and rhythm." Health is another domain where GPT-5 excels. In health-related questions, GPT-5 scores significantly higher than any previous model. Now coming to benchmarks, in SWE-bench Verified which evaluates software engineering tasks, GPT-5 achieved 74.9% accuracy with thinking. In Aider Polyglot (code editing benchmark), GPT-5 scored 88% with thinking (pass@2). Next, in the challenging Humanity's Last Exam test, GPT-5 Pro scored a whopping 42% with tools access. And in the GPQA Diamond benchmark (PhD-level science questions), GPT-5 pro scored 89.4% with tools. In the Tau2-bench agentic benchmark, GPT-5 achieved 96.7% in telecom, 81.1% in retail, and 62.6% in airline. In the MMMU benchmark which tests visual problem-solving, GPT-5 got 84.2% with thinking. What is interesting about GPT-5 is that it hallucinates much less than previous OpenAI models. Its hallucination rate is 0.7% in LongFace-Concepts, compared to o3's 4.5%. GPT-5 is rolling out to free ChatGPT users, starting today, with a limited volume of queries. If free users reach their usage limit, they can still use the smaller GPT-5 mini model. Note that the full reasoning capability of GPT-5 will be slowly rolled out to free users. In addition, GPT-5 is the new default model on ChatGPT for all users. It's replacing GPT‑4o, OpenAI o3, OpenAI o4-mini, GPT‑4.1, and GPT‑4.5 for all signed-in users. As for ChatGPT Plus users ($20 per month), they get access to the GPT-5 model with reasoning capability, and with much higher limits. ChatGPT Pro users who pay $200 per month and Team users will have access to the GPT-5 Pro model. It uses more compute and thinking time to solve challenging problems. OpenAI says Enterprise and Edu users will get access to GPT-5 in one week. On the API side for developers, GPT-5 offers a context window of 400K tokens and can output up to 128K tokens. GPT-5's knowledge cutoff is October o1, 2024. Finally, GPT-5's API pricing is $1.25/$10 for input/output per 1 million token.
[64]
OpenAI Rolls Out GPT-5 To All ChatGPT Users As Sam Altman Hails It The Best Model Ever For Writing, Coding, And Human-Level Reasoning
OpenAI has officially rolled out its most advanced AI model yet, the GPT-5, after a year of building up its hype. It is being made available to all ChatGPT users and developers. CEO Sam Altman compared the debut of the model to the moment the first iPhone came out with the Retina display - an experience that, once you indulge in it, you can never go back on. The next-generation language model brings some major improvements in reasoning, accuracy, and speed, and brings interactions that are closer to engaging with a highly trained expert. One of the most significant improvements in the GPT-5 model is the more simplified user experience it brings about, as instead of juggling between multiple models, GPT-5 operates as a fully unified system that uses an intelligent router to automatically switch to more advanced reasoning when presented with complex prompts, eliminating the constant switching between separate models. For developers, OpenAI brings about three options for accessing the API: GPT-5, GPT-5 Mini, and GPT-5 Nano, and each one is optimized for different resource needs and varied performance. The model's domain-specific capabilities also tend to outshine, and Altman is asserting that the GPT-5 is the best model in the world, specifically for writing, coding, and healthcare. Greater personalization is added as users can customize the experience by choosing the response style with options like "Listener" or even "Cynic" available, and you can even change the color of the chat. There are also enhanced safety features added that help ensure the responses are measured and cautious. Instead of refusing to answer uncertain questions, it would now be offering what is called safe completions, which ensure the information needed is provided without overstepping. This shows how OpenAI is working vigorously to bring more usefulness, yet not compromising on offering responsible AI. OpenAI is quite literally raising the AI bar with the GPT-5 model as it tends to retain the capabilities of GPT-4o, but makes it even better with a faster and more responsive experience. It can easily understand multiple formats, making it a robust assistant across different types of tasks. One of the surprising parts of the announcement is that the new model would be available for free. ChatGPT free-tier users can use the model initially, but after they hit the usage cap, they will be switched to the GPT-5 Mini, which is a lighter version of the model. The paid users will have access to the full experience of GPT-5. It will be available starting today for the multiple-tier users and is one of the company's most inclusive rollouts. From a broader perspective, the launch of the GPT-5 model feels like a step towards Artificial General Intelligence (AGI), which Altman has hinted previously to be reachable by 2027. Although it is not AGI itself, it is laying the infrastructure for models that will come with general, human-like intelligence in the future.
[65]
OpenAI Launches GPT-5 Amid Competition From Elon Musk's Grok
GPT-5 to support 700 million users, fuels $500 billion OpenAI valuation. On Thursday, August 7, 2025, OpenAI will reveal its latest and most powerful language model, GPT-5 at an online presentation. The release comes as pressure mounts from Elon Musk's xAI and its chatbot Grok. GPT-5 Launch Brings Expert-Level AI, Custom Personalities, and Instant App Creation According to a Bloomberg report, GPT-5 will become available to free users and most paid ChatGPT subscribers on launch day. Enterprise and education customers will gain access the following week. The new model delivers faster responses, stronger reasoning, and more accurate outputs than earlier versions. CEO Sam Altman described GPT-5 as a major upgrade, saying it now feels like speaking to an expert on any topic. He shared that the model recently solved a complex question that he couldn't answer himself. ChatGPT will offer four built-in personalities with the update (Cynic, Robot, Listener, and Nerd) designed to customize user interactions. GPT-5 also decides automatically how deeply to process each prompt, eliminating the need for manual settings. OpenAI's launch comes amid rising competition. Musk's xAI is positioning Grok as a serious contender. The model was recently tasked with exploring whether quantum computing poses a real threat to Bitcoin. This shows xAI's aim to enter technical and financial domains once led by OpenAI. GPT-5 is also designed to improve software development. Altman said this signals the rise of "software on demand," where applications can be created instantly using AI. During internal tests, OpenAI's latest model built a fully functional French language-learning app with games and quizzes in minutes. Early testers, including Anysphere and Lovable, found the model faster and more reliable. Musk Claims Grok Surpasses OpenAI's Latest Model as AI Race Intensifies Meanwhile, Elon Musk has claimed that Grok 4 Heavy was already smarter than GPT-5 two weeks ago. In a recent post, Musk said Grok 4 Heavy is "a lot better." His comment followed a benchmark showing GPT-5 with tools ranks between Grok 4 and Grok 4 Heavy. However, Lovable's CEO said GPT-5 handled complex apps better than competitors and produced clean, maintainable code while fixing bugs effectively. To support GPT-5 and future growth, OpenAI has raised tens of billions and holds a $300 billion valuation. It is reportedly in talks for a secondary stock sale that could value the company at $500 billion. ChatGPT now serves 700 million users weekly and five million business clients. OpenAI is offering U.S. federal agencies access for just $1 per year.
[66]
OpenAI launches GPT-5 as the AI industry seeks a return on investment
SAN FRANCISCO, Aug 7 (Reuters) - OpenAI launched on Thursday its GPT-5 artificial intelligence model, the highly anticipated latest installment of a technology that has helped transform global business and culture. OpenAI's GPT models are the AI technology that powers the popular ChatGPT chatbot, and GPT-5 will be available to all 700 million ChatGPT users, OpenAI said. The big question is whether the company that kicked off the generative AI frenzy will be capable of continuing to drive significant technological advancements that attract enterprise-level users to justify the enormous sums of money it is investing to fuel these developments. The release comes at a critical time for the AI industry. The world's biggest AI developers - Alphabet, Meta, Amazon and Microsoft, which backs OpenAI - have dramatically increased capital expenditures to pay for AI data centers, nourishing investor hopes for great returns. These four companies expect to spend nearly $400 billion this fiscal year in total. OpenAI is now in early discussions to allow employees to cash out at a $500 billion valuation, a huge step-up from its current $300 billion valuation. Top AI researchers now command $100 million signing bonuses. "So far, business spending on AI has been pretty weak, while consumer spending on AI has been fairly robust because people love to chat with ChatGPT," said economics writer Noah Smith. "But the consumer spending on AI just isn't going to be nearly enough to justify all the money that is being spent on AI data centers." OpenAI is emphasizing GPT-5's enterprise prowess. In addition to software development, the company said GPT-5 excels in writing, health-related queries, and finance. "GPT-5 is really the first time that I think one of our mainline models has felt like you can ask a legitimate expert, a PhD-level expert, anything," OpenAI CEO Sam Altman said at a press briefing. "One of the coolest things it can do is write you good instantaneous software. This idea of software on demand is going to be one of the defining features of the GPT-5 era." One key measure of success is whether the step up from GPT-4 to GPT-5 is on par with the research lab's previous improvements. Two early reviewers told Reuters that while the new model impressed them with its ability to code and solve science and math problems, they believe the leap from the GPT-4 to GPT-5 was not as large as OpenAI's prior improvements. Nearly three years ago, ChatGPT introduced the world to generative AI, dazzling users with its ability to write humanlike prose and poetry, quickly becoming one of the fastest growing apps ever. In March 2023, OpenAI followed up ChatGPT with the release of GPT-4, a large language model that made huge leaps forward in intelligence. While GPT-3.5, an earlier version, received a bar exam score in the bottom 10%, GPT-4 passed, opens new tab the simulated bar exam in the top 10%. GPT-4's leap was based on more compute power and data, and the company was hoping that "scaling up" in a similar way would consistently lead to improved AI models. But OpenAI ran into issues scaling up. One problem was the data wall the company ran into, and OpenAI's former chief scientist Ilya Sutskever said last year that while processing power was growing, the amount of data was not. He was referring to the fact that large language models are trained on massive datasets that scrape the entire internet, and AI labs have no other options for large troves of human-generated textual data. Apart from the lack of data, another problem was that 'training runs' for large models are more likely to have hardware-induced failures given how complicated the system is, and researchers may not know the eventual performance of the models until the end of the run, which can take months. At the same time, OpenAI discovered another route to smarter AI, called "test-time compute," a way to have the AI model spend more time compute power "thinking" about each question, allowing it to solve challenging tasks such as math or complex operations that demand advanced reasoning and decision-making. GPT-5 acts as a router, meaning if a user asks GPT-5 a particularly hard problem, it will use test-time compute to answer the question. This is the first time the general public will have access to OpenAI's test-time compute technology, something that Altman said is important to the company's mission to build AI that benefits all of humanity. Altman believes the current investment in AI is still inadequate. "We need to build a lot more infrastructure globally to have AI locally available in all these markets," Altman said.
[67]
GPT-5's Personalities and why they work
Switching between GPT-5 personas creates smoother, more natural conversations for varied workflows When GPT-5 rolled out with its new "personalities" feature, I didn't expect to spend much time on it. My plan was to quickly test it, maybe switch modes a few times, and then move on. But there's something strangely intriguing about being able to flip your AI between four distinct personas - Cynic, Robot, Listener, and Nerd. I haven't used them enough to call myself an expert, but even in these first hours, I can see why this update matters. It's not just a new skin for the same brain, it's a shift in how the brain itself behaves. Also read: GPT-5 launched: Sam Altman's 3 key claims on AI, AGI and India OpenAI's GPT-5 personalities are designed to give you four ready-made styles of interaction without the need for elaborate prompting. The first thing I noticed is that these aren't half-hearted tone changes, they feel like different people handling the same conversation. The interesting thing is how quickly you can jump between them without losing the thread of the conversation. One second you're getting blunt criticism from the Cynic, and the next you're having the same idea warmly encouraged by the Listener. Normally, getting an AI to match your tone requires constant prompting - "Be concise," "Add humor," "Be critical" - and that instruction can wear off mid-conversation. GPT-5's personalities make that process instant and persistent. Also read: Soon after Sam Altman unveils GPT-5, Elon Musk says Grok-5 will be out this year as well Even with limited use, I can see the value in having a consistent style locked in from the moment you start. If you're drafting a technical guide, Robot mode won't suddenly start cracking jokes halfway through. If you're brainstorming with Listener, it won't suddenly turn into a hard-nosed fact-checker unless you tell it to. There's also a deeper reason this works: humans adapt their speech depending on the person they're talking to. These personalities give GPT-5 that same adaptability, making it feel more natural. Instead of one "default" assistant, you have four conversational specialists at your fingertips. Steerability, the ability to guide an AI's style and behavior, isn't new, but GPT-5's take on it is much cleaner. Switching between personalities mid-discussion doesn't just swap vocabulary; it changes how the model approaches the problem. The Cynic will look for flaws, the Robot will look for precision, the Listener will focus on your intent, and the Nerd will hunt for interesting tangents. Even without hours of testing, it's easy to imagine how this could reshape workflows. A quick content outline could be roughed out in Nerd mode, critiqued in Cynic mode, fact-checked in Robot mode, and polished with the human touch of Listener mode, all without leaving the same chat window. If this is just the launch version, the possibilities ahead are much bigger. Right now, the four personalities feel like proof-of-concepts: a starting set to show that tone can be swapped instantly without breaking reasoning quality. But what if we go beyond tone? Imagine profession-specific personas: a legal expert who explains clauses in plain English, a project manager who keeps track of your deadlines, or a tutor who adapts to your learning pace. The fact that GPT-5's current personalities already change how it thinks, not just how it speaks, suggests we're heading in that direction. I've not had enough time with GPT-5's personalities, so I can't yet say which will become my go-to. But even in these early interactions, the difference is obvious. This isn't just a voice change, it's a mindset change. Each personality shapes the flow of conversation, the kind of answers you get, and the way you respond in turn. Right now, I'm still exploring, still figuring out which mode fits which task. But I can already see the appeal. Sometimes you need encouragement, sometimes precision, sometimes a challenge, and now, GPT-5 can switch between those roles on command. If this is what day one feels like, I'm curious to see how much more natural and indispensable these personalities will become once they've been part of my routine for weeks instead of hours.
[68]
OpenAI launches GPT-5, its most advanced model yet: What's new, what it can do, and who gets access
Rolling out to Free, Plus, Pro, and Team users now; Enterprise and Edu users will get access in a week. OpenAI has officially launched GPT-5, calling it their most potent and capable model to date. The new model is available as the default in ChatGPT, replacing GPT‑4o, OpenAI o3, OpenAI o4-mini, GPT‑4.1, and GPT‑4.5 for signed-in users. It is claimed to outperform previous models across industry benchmarks, scoring 74.9% on SWE-bench Verified and a record 88% on Aider polyglot, which measures code editing accuracy. OpenAI, in its blog post, said that GPT-5 is released in three sizes: GPT-5, GPT-5-mini, and GPT-5-nano, giving users the ability to balance performance, cost, and speed. Here's everything you need to know. GPT-5 provides a quick answer, depending on the complexity of your question. It's also noticeably better at writing, answering real-world questions, and especially coding. It can build websites, apps, and even games in a single prompt, with cleaner code and a better sense of design. It's also more accurate and more responsive. Additionally, it's also their most advanced health assistant yet. GPT-5 can deliver significantly more accurate and context-aware responses to health-related questions. While it doesn't replace a doctor, it can help users better understand medical information, ask more thoughtful questions, and make informed decisions. Further, it is said to set a new state of the art across math (94.6% on AIME 2025 without tools), real-world coding (74.9% on SWE-bench Verified, 88% on Aider Polyglot), multimodal understanding (84.2% on MMMU), and health (46.2% on HealthBench Hard). Also read: Bengaluru man duped of Rs 1.77 crore in digital arrest scam: Here's what happened GPT-5 will also allow users to pick a custom chat colour, with exclusive shades unlocked for Pro subscribers. It is also getting four built-in personalities- Cynic, Robot, Listener, and Nerd, letting users prefer how the AI responds - sarcastic, supportive, or straightforward. For productivity, Pro users will soon be able to connect Gmail, Google Calendar, and Google Contacts directly to ChatGPT. Voice interactions are also getting smarter. The new Advanced Voice feature can better understand users' tone and adapt its responses in real-time. Though it's available to all users, higher usage limits are available for paid plans. OpenAI is also retiring the older Standard Voice Mode in 30 days. GPT‑5 rolling out to all Plus, Pro, Team, and Free users starting today. Access for Enterprise and Edu will be made available in one week. Pro, Plus, and Team users can also do coding with GPT‑5 in the Codex CLI by simply signing in to ChatGPT.
Share
Copy Link
OpenAI releases GPT-5, its latest AI model, offering improved reasoning, coding capabilities, and accessibility to all ChatGPT users, including those on the free tier.
OpenAI has launched GPT-5, its latest flagship AI model, marking a significant milestone in the company's pursuit of artificial general intelligence (AGI). The new model, released on Thursday, is now available to all ChatGPT users, including those on the free tier, representing a departure from OpenAI's previous strategy of gating advanced models behind a paywall 123.
Source: Analytics India Magazine
GPT-5 is touted as OpenAI's "best AI system yet," offering a range of improvements over its predecessors:
Reduced Confabulations: The model demonstrates a significant reduction in hallucinations, with OpenAI reporting that GPT-5 (with thinking) hallucinates only 4.5% of the time, compared to 22% for o3 and 20.5% for GPT-4o 3.
Improved Coding Abilities: GPT-5 excels in coding tasks, scoring 74.9% on SWE-bench Verified, slightly outperforming Anthropic's Claude Opus 4.0 (74.5%) and Google DeepMind's Gemini 2.0 Pro (59.5%) 3.
Source: CNET
GPT-5 represents OpenAI's attempt to unify various AI capabilities into a single system:
Real-time Router: The model includes a smart router that decides which approach to use based on conversation type, complexity, tool needs, and user intent 13.
Multimodal Interactions: Like its predecessor GPT-4o, GPT-5 can interact via images, voice, and text 1.
Improved User Interface: OpenAI has focused on making ChatGPT simpler to use, with the model automatically choosing whether to apply reasoning to each query 4.
OpenAI has implemented new safety features in GPT-5:
Safe Completions: A new technique for handling ambiguous queries that may or may not be problematic, aiming to provide helpful answers within safety constraints 2.
Improved Health-related Responses: GPT-5 is reported to be more accurate and proactive in flagging potential health concerns and helping users parse medical results 3.
The release of GPT-5 is expected to have significant implications for the AI industry:
User Base: ChatGPT currently boasts over 700 million weekly active users, nearly 10% of the global population 3.
Competitive Landscape: While GPT-5 outperforms competitors in some areas, it slightly underperforms in others, maintaining a competitive but not dominant position in the market 34.
Source: Economic Times
As the AI landscape continues to evolve rapidly, the introduction of GPT-5 represents another step forward in the development of more capable and accessible AI systems. However, questions remain about the true extent of its advancements and its potential impact on various industries and society as a whole 5.
NVIDIA CEO Jensen Huang confirms the development of the company's most advanced AI architecture, 'Rubin', with six new chips currently in trial production at TSMC.
2 Sources
Technology
22 hrs ago
2 Sources
Technology
22 hrs ago
Databricks, a leading data and AI company, is set to acquire machine learning startup Tecton to bolster its AI agent offerings. This strategic move aims to improve real-time data processing and expand Databricks' suite of AI tools for enterprise customers.
3 Sources
Technology
22 hrs ago
3 Sources
Technology
22 hrs ago
Google is providing free users of its Gemini app temporary access to the Veo 3 AI video generation tool, typically reserved for paying subscribers, for a limited time this weekend.
3 Sources
Technology
14 hrs ago
3 Sources
Technology
14 hrs ago
Broadcom's stock rises as the company capitalizes on the AI boom, driven by massive investments from tech giants in data infrastructure. The chipmaker faces both opportunities and challenges in this rapidly evolving landscape.
2 Sources
Technology
22 hrs ago
2 Sources
Technology
22 hrs ago
Apple is set to introduce new enterprise-focused AI tools, including ChatGPT configuration options and potential support for other AI providers, as part of its upcoming software updates.
2 Sources
Technology
22 hrs ago
2 Sources
Technology
22 hrs ago