OpenAI Exec Exodus: COO and AGI CEO Step Aside Before IPO
OpenAI's COO, AGI CEO Fidji Simo, and two other executives stepped aside weeks before a potential IPO. Codex is now the foundation for everything.
OpenAI Kills Its Erotic Chatbot: Even AI Has Lines It Won't Cross
OpenAI has shelved plans to launch an erotic ChatGPT mode, backing away from a controversial expansion that would have allowed adult users to generate sexual content. Internal concerns about societal impact drove the reversal.
OpenAI Raised $122B. Investors Are Dumping the Shares.
OpenAI just closed the largest venture round in history. But institutional holders can't unload $600M in stock. The valuation paradox reshaping AI.
Nvidia's $2B Marvell Bet and the AI Networking War
Nvidia pours $2 billion into Marvell to dominate AI networking. Nothing plans smart glasses for 2027. AI capex hits $635B but energy costs surge.
Ethereum Foundation Stakes $93M in ETH в— A Bold Vote of Confidence in the Network
The Ethereum Foundation deposited 45,034 ETH to the Beacon Chain, bringing its total staked position to nearly 69,500 ETH worth $143 million.
Crypto's Q1 Bloodbath: Bitcoin Drops 22%, But Infra Builds On
Bitcoin shed a quarter of its value in Q1 2026 в— its worst quarter since 2018. Miners sold $250M in BTC. Ethereum Foundation staked $93M. The divergence signals a maturing market.
Claude Mythos Leaked: Anthropic's Secret 'Most Capable' Model Exposed Online
Unpublished files describing Anthropic's Claude Mythos, internally dubbed the company's most capable AI model, leaked online this week, raising questions about internal security and the competitive AI model race.
Claude Code's Source Leaked. The Web Is Now an Attack Surface.
Anthropic's Claude Code source was leaked via npm. Google warns the open web is the #1 attack vector for AI agents. Prompt injections hit 86% success.
China's AI Spies: First Documented Autonomous Cyber Espionage Campaign Using Claude Code
Chinese state hackers weaponized Claude Code to execute 80-90% of cyber operations autonomously against 30 global targets — the first documented AI-driven espionage campaign of its kind.
Brain-Inspired Chip Promises 2000x AI Energy Efficiency Breakthrough
Researchers at Loughborough University built a brain-inspired chip that processes time-series data directly in hardware, slashing energy use by up to 2000x compared to traditional AI systems.
Baidu's Robotaxis Froze En Masse in Wuhan — Stranding Passengers on Highways
A system failure hit over 100 of Baidu's Apollo Go robotaxis in Wuhan, stopping them mid-traffic, stranding riders, and sparking fresh safety concerns for autonomous vehicles.
ARC-AGI-3 Benchmark Shatters AGI Hype: Top Models Score Below 1%
The ARC-AGI-3 benchmark tested whether AI agents can generalize in unknown environments. Top models scored under 1% while humans solved 100%. The verdict: AGI is nowhere near here.
Anthropic Locks the Garden: Claude Subscriptions Cut Off from OpenClaw and Third-Party Agents
Anthropic ends the era of flat-rate Claude subscriptions powering third-party AI agents, forcing power users into pay-as-you-go billing.
Anthropic Forms AnthroPAC — The AI Company Goes Political
Anthropic files with the FEC to launch a political action committee, escalating the AI policy wars as the company battles the Trump administration in federal court.
Anthropic Acquires AI Biotech Startup Coefficient Bio for $400M
Anthropic is pushing beyond general-purpose AI with its $400M acquisition of Coefficient Bio, a startup applying AI to drug discovery and clinical workflows. The move signals AI's expansion into high-stakes scientific domains.