LLMs

Twitter

Facebook

Copy Link

Key Facts

Year2018

Key PlayersOpenAI, Google, Anthropic

Top News

Tech Giants Slam the Brakes on AI Spending as Tokenmaxxing Drains Budgets Without Results

Major companies including Uber, Amazon, and Microsoft are pulling back from widespread AI use as costs spiral out of control. After Uber exhausted its entire 2026 Claude Code budget by April, executives admit they can't draw a clear line between rising token consumption and useful consumer features. Amazon shut down its internal AI leaderboard after employees engaged in tokenmaxxing, while one unidentified company reportedly burned through $500 million in a single month after failing to set usage limits.

Business16 sources

Tech Giants Slam the Brakes on AI Spending as Tokenmaxxing Drains Budgets Without Results

Business16 sources

AI in Legal Proceedings Fuels Sharp Rise in Pro Se Litigants, Straining US Courts

Self-represented litigants using AI tools like ChatGPT and Claude are filing lawsuits at unprecedented rates, with pro se cases jumping from 11% to 16.8% of federal civil cases. While AI chatbots for legal documents promise better access to justice for those without legal representation, federal courts face mounting strain from AI-generated lawsuits that often contain hallucinated case law and meritless filings.

Entertainment5 sources

AI in Legal Proceedings Fuels Sharp Rise in Pro Se Litigants, Straining US Courts

Entertainment5 sources

DeepSeek makes 75% price cut permanent, intensifying AI price war with OpenAI and Anthropic

Chinese AI startup DeepSeek has permanently slashed prices on its flagship V4 Pro model by 75%, bringing costs down to $0.87 per million output tokens. The aggressive move undercuts OpenAI's GPT-5 and Anthropic's Claude by up to 34x, forcing a reckoning in the enterprise AI market as token costs become the binding constraint on adoption.

Technology5 sources

DeepSeek makes 75% price cut permanent, intensifying AI price war with OpenAI and Anthropic

Technology5 sources

Alibaba's Qwen3.7-Max AI model runs 35 hours autonomously, outperforms rivals on coding benchmarks

Alibaba launched Qwen3.7-Max, a proprietary AI model capable of 35 hours of continuous autonomous execution on complex engineering tasks. The model scored 60.6 on Swaybench for long-term coding tasks, surpassing competitors like Claude Opus-4.6 and DeepSeek V4-Pro. It supports external frameworks including Anthropic's Claude Code and offers a 1-million-token context window.

Technology2 sources

Alibaba's Qwen3.7-Max AI model runs 35 hours autonomously, outperforms rivals on coding benchmarks

Technology2 sources

AMD prices Ryzen AI Halo at $3,999 with June preorders, challenges Nvidia's DGX Spark dominance

AMD unveiled pricing for its Ryzen AI Halo developer platform at $3,999, undercutting Nvidia's DGX Spark by $700. The compact AI workstation features 128GB of unified memory and supports both Windows and Linux, targeting developers seeking cost-effective alternatives to cloud-based AI solutions. Preorders begin in June 2026.

Technology15 sources

AMD prices Ryzen AI Halo at $3,999 with June preorders, challenges Nvidia's DGX Spark dominance

Technology15 sources

Alibaba reveals Zhenwu M890 AI chip but admits struggling with production scale

Alibaba launched the Zhenwu M890 AI chip, claiming triple the performance of its predecessor, but disclosed it has only shipped 560,000 units total—a fraction of Nvidia's output. The announcement highlights China's push for semiconductor self-reliance amid US export restrictions, while exposing the production challenges Chinese chipmakers face without access to advanced manufacturing processes.

Technology8 sources

Alibaba reveals Zhenwu M890 AI chip but admits struggling with production scale

Technology8 sources

AI passes the Turing Test as GPT-4.5 appears more human than actual people in landmark study

A University of California San Diego study provides the first empirical evidence that AI passes the Turing Test. GPT-4.5 was judged human 73% of the time—more often than actual humans—fundamentally changing how we measure machine intelligence. The findings raise urgent questions about online trust and distinguishing between humans and AI in everyday digital interactions.

Science4 sources

AI passes the Turing Test as GPT-4.5 appears more human than actual people in landmark study

Science4 sources

Baidu AI becomes core growth engine as cloud revenue surges 79% despite profit slide

Baidu reported first-quarter revenue of $4.65 billion, narrowly beating estimates as AI cloud infrastructure climbed 79% year-over-year. CEO Robin Li declared AI as the company's primary growth engine after AI-powered business exceeded half of general business revenue for the first time. However, net profit fell 55% amid ongoing investments in autonomous driving and chip development.

Business3 sources

Baidu AI becomes core growth engine as cloud revenue surges 79% despite profit slide

Business3 sources

AI Grade Inflation Surges 30% at Universities as Students Learning Less, Study Reveals

A UC Berkeley study analyzing over 500,000 student enrollments reveals that AI sends A grades into overdrive, with excellent grades rising 30% in AI-exposed courses since ChatGPT's release. While students achieve higher GPAs through AI-assisted cheating, researchers warn this trend creates a less competent workforce unable to perform core job duties.

Entertainment2 sources

AI Grade Inflation Surges 30% at Universities as Students Learning Less, Study Reveals

Entertainment2 sources

Amazon Games forced team to pivot Project Trident to AI, then laid them off anyway

Amazon Game Studios pressured developers to abandon their Shadow of the Colossus-style game to build Project Trident with generative AI integration. The team scrambled through multiple pivots over 18 months, creating a Viking-inspired action comedy with LLM-powered NPCs. Despite nearing a playable demo, Amazon laid off the entire team in October 2025 alongside 14,000 employees.

Business3 sources

Amazon Games forced team to pivot Project Trident to AI, then laid them off anyway

Business3 sources

Claude AI helps recover $400,000 Bitcoin wallet locked for 11 years after password mishap

A Bitcoin holder who forgot their wallet password while high in college has finally regained access to nearly $400,000 worth of cryptocurrency with help from Anthropic's Claude AI. After 11 years of failed attempts and 3.5 trillion password combinations, the AI discovered an old wallet backup file and identified a critical bug that had prevented recovery all along.

Technology12 sources

Claude AI helps recover $400,000 Bitcoin wallet locked for 11 years after password mishap

Technology12 sources

State media control shapes what AI chatbots say by flooding training data with biased content

A new Nature study reveals governments indirectly influence large language models by controlling online media environments. Researchers found state-coordinated media appears 41 times more frequently than Wikipedia in Chinese training data, causing ChatGPT and Claude to produce more pro-government responses when prompted in native languages versus English.

Science3 sources

State media control shapes what AI chatbots say by flooding training data with biased content

Science3 sources

European Central Bank warns euro area banks to act now on AI cyber threats from Mythos

The European Central Bank's Frank Elderson is urging euro area banks to immediately prepare for potential cyberattacks powered by Anthropic's Mythos AI model. Despite lacking access to the tool, Elderson emphasized that inaction is not an option as banks must brace for increasingly aggressive AI-driven cyber warfare. Large U.S. banks with early Mythos access are already rushing to fix data system vulnerabilities.

Policy2 sources

European Central Bank warns euro area banks to act now on AI cyber threats from Mythos

Policy2 sources

Princeton University ends 133-year-old tradition as AI cheating forces return of exam proctors

Princeton University has abandoned its 133-year-old Honor Code tradition of proctor-free exams, marking one of the most significant policy reversals in higher education. The change comes after surveys revealed 30% of seniors admitted to cheating, with generative AI tools making academic dishonesty easier than ever. Starting July 1, human proctors will monitor all exams at the Ivy League institution.

Entertainment5 sources

Princeton University ends 133-year-old tradition as AI cheating forces return of exam proctors

Entertainment5 sources

Sipeed K3 RISC-V SBC runs 30B AI LLMs locally at 15 tokens per second for $299

Sipeed has launched its K3 series RISC-V-powered single board computers that can run 30B-parameter AI LLMs locally at up to 15 tokens per second. Built with SpacemiT's Fusion Architecture and featuring up to 32GB LPDDR5 memory plus a 60 TOPS NPU, the K3 offers an accessible open-source alternative to proprietary AI hardware starting at $299.

Technology2 sources

Sipeed K3 RISC-V SBC runs 30B AI LLMs locally at 15 tokens per second for $299

Technology2 sources

Prev 1…789…27 Next