LLMs

Twitter

Facebook

Copy Link

Key Facts

Year2018

Key PlayersOpenAI, Google, Anthropic

Top News

DuckDuckGo Launches Private AI Voice Chat on Duck.ai With Encrypted Audio Streaming

DuckDuckGo has rolled out AI voice chat capabilities for Duck.ai, its private AI chatbot platform. The feature enables real-time voice conversations with large language models while maintaining strict privacy protections. Audio streams are encrypted and not stored by DuckDuckGo or OpenAI, addressing growing concerns about data collection in AI applications.

Technology3 sources

DuckDuckGo Launches Private AI Voice Chat on Duck.ai With Encrypted Audio Streaming

Technology3 sources

Vitalik Buterin Rejects 'Race to AGI,' Proposes Ethereum-Led Alternative for Safer AI Development

Ethereum co-founder Vitalik Buterin is challenging Silicon Valley's AGI acceleration narrative, calling the 'race to AGI' a flawed framework that prioritizes status over safety. In a detailed post on X, he outlined a four-part roadmap positioning Ethereum as infrastructure for private, verifiable AI interactions, an economic layer for AI agents, and enhanced governance systems—offering a decentralized alternative to centralized AI development.

Technology4 sources

Vitalik Buterin Rejects 'Race to AGI,' Proposes Ethereum-Led Alternative for Safer AI Development

Technology4 sources

AI chatbots fail to improve medical advice for patients, Oxford study reveals

A University of Oxford study published in Nature Medicine found that AI chatbots offer no advantage over internet searches when patients seek medical advice. Despite large language models achieving 94.9% accuracy in controlled tests, real-world human-AI interaction in healthcare revealed a troubling gap between AI potential and performance, with patients struggling to provide complete information and receiving inconsistent advice.

Health14 sources

AI chatbots fail to improve medical advice for patients, Oxford study reveals

Health14 sources

Apple's Siri AI upgrade hits another snag, pushing key features to late 2026 or beyond

Apple's long-awaited Siri overhaul faces yet another setback as internal testing reveals the revamped virtual assistant struggles with speed and accuracy. Originally announced in 2024, the AI-powered upgrade—built on Google Gemini integration—was slated for March 2026 but will now roll out piecemeal across future iOS updates, potentially delaying some features until September or later.

Technology24 sources

Apple's Siri AI upgrade hits another snag, pushing key features to late 2026 or beyond

Technology24 sources

Anthropic's Claude Opus 4.6 finds 500+ security flaws, sparking dual-use concerns

Anthropic released Claude Opus 4.6, which discovered over 500 previously unknown high-severity security flaws in open-source libraries during testing. The large language model identified zero-day vulnerabilities without specialized prompting, using human-like reasoning to spot patterns in codebases. While the capability strengthens defenders, it also raises cybersecurity risks as the same technology could be weaponized by attackers.

Technology2 sources

Anthropic's Claude Opus 4.6 finds 500+ security flaws, sparking dual-use concerns

Technology2 sources

Claude Sonnet 5 leak reveals one-million-token context and imminent Anthropic release

Anthropic is preparing to release Claude Sonnet 5, a new mid-tier AI model that could reshape the large language model market. Leaked references found in Google Vertex AI logs point to a February 3, 2026 launch, with the model codenamed Fennec showing a one-million-token context window, 50% lower cost than Opus 4.5, and TPU-based training for faster inference speeds.

Technology3 sources

Claude Sonnet 5 leak reveals one-million-token context and imminent Anthropic release

Technology3 sources

Microsoft builds scanner to detect hidden backdoors in AI models using three warning signs

Microsoft's AI Security team developed a lightweight scanner that identifies backdoors in open-weight large language models. The tool detects model poisoning by analyzing three behavioral signals: abnormal attention patterns, memorization of poisoned data, and fuzzy trigger responses. Unlike traditional malware, these sleeper agents remain dormant until activated by specific trigger phrases, making them extremely difficult to detect through standard safety testing.

Technology4 sources

Microsoft builds scanner to detect hidden backdoors in AI models using three warning signs

Technology4 sources

Sony patents AI-generated podcasts hosted by PlayStation characters like Kratos and Aloy

Sony has secured a patent for AI-generated podcasts that would feature voices of beloved PlayStation characters delivering personalized game news, trophy updates, and recommendations. The patent, titled 'LLM-Based Generative Podcasts for Gamers,' aims to address what Sony calls a gap in providing targeted content to gamers on its platform.

Technology7 sources

Sony patents AI-generated podcasts hosted by PlayStation characters like Kratos and Aloy

Technology7 sources

NASA's Perseverance Rover Completes First AI-Planned Drive on Mars Using Claude AI Models

NASA's Perseverance rover achieved a milestone by completing the first drives on another planet planned entirely by artificial intelligence. On December 8 and 10, the rover traveled nearly 1,500 feet across Mars using routes generated by Anthropic's Claude AI models, marking a shift from decades of manual route planning by human drivers at the Jet Propulsion Laboratory.

Technology11 sources

NASA's Perseverance Rover Completes First AI-Planned Drive on Mars Using Claude AI Models

Technology11 sources

OpenAI Retires GPT-4o Model on Feb 13, Pushing Users to GPT-5.2 Despite Loyal Following

OpenAI announced it will retire GPT-4o and several other ChatGPT models on February 13, just two weeks after the announcement. Despite the model's beloved conversational style and warmth, OpenAI says only 0.1% of users still choose it daily, representing about 800,000 people. The company previously reversed this decision after intense user backlash but now feels confident that GPT-5.2 and new personalization features can satisfy longtime fans.

Technology13 sources

OpenAI Retires GPT-4o Model on Feb 13, Pushing Users to GPT-5.2 Despite Loyal Following

Technology13 sources

AI pioneer Yann LeCun warns business tools built on LLMs are headed for obsolescence

Turing Award winner Yann LeCun is sounding the alarm on Silicon Valley's AI strategy. After leaving Meta to launch his own startup, the renowned researcher argues that large language models represent a dead end for achieving human-level intelligence. He warns that companies are wasting billions on limited technology while Chinese competitors pursue more promising approaches.

Technology2 sources

AI pioneer Yann LeCun warns business tools built on LLMs are headed for obsolescence

Technology2 sources

CISA's Acting Director Uploaded Sensitive Files to ChatGPT, Triggering Federal Investigation

Dr. Madhu Gottumukkala, acting director of CISA, uploaded sensitive government documents into the public version of ChatGPT last summer, triggering automated security alerts. The incident has sparked a Department of Homeland Security investigation into potential data exposure and raises critical questions about federal AI governance and human element vulnerability in cybersecurity.

Policy2 sources

CISA's Acting Director Uploaded Sensitive Files to ChatGPT, Triggering Federal Investigation

Policy2 sources

Researchers develop framework to give AI metacognition and self-awareness capabilities

Scientists have created a mathematical framework that enables generative AI systems like ChatGPT to monitor their own cognitive processes through metacognition. The system uses a five-dimensional metacognitive state vector to help AI assess confidence, detect confusion, and decide when problems require deeper analysis, potentially transforming high-stakes applications.

Science2 sources

Researchers develop framework to give AI metacognition and self-awareness capabilities

Science2 sources

Meta stock surges 10% as revamped AI strategy delivers strong fourth-quarter financial results

Meta exceeded Wall Street expectations with $59.89 billion in Q4 revenue, up 24% year-over-year, and earnings of $8.88 per share. The company's artificial intelligence investments are paying off with measurable improvements in ad performance and engineering productivity. Despite projecting capital expenditures of $115-135 billion for 2026, investors rallied behind Meta's bullish revenue guidance and evidence that its revamped AI strategy is delivering tangible returns.

Business10 sources

Meta stock surges 10% as revamped AI strategy delivers strong fourth-quarter financial results

Business10 sources

OpenAI launches dedicated team to position AI as scientific collaborator across research fields

OpenAI has established a new AI for Science team led by Kevin Weil to help scientists accelerate research across mathematics, physics, chemistry, and biology. The company reports that 1.3 million weekly users now send 8.4 million messages on advanced science topics, representing 47% growth over the past year. GPT-5.2 achieves 92% accuracy on graduate-level benchmarks, though questions remain about long-term validation.

Technology4 sources

OpenAI launches dedicated team to position AI as scientific collaborator across research fields

Technology4 sources

Prev 1…121314…22 Next