Topic
27 stories from Planck Standard. Latest: September 21, 2026.
WatchThe Saturday Truth Social post gave no structure, budget, authority or name for the new body. It landed a week after Anthropic's Dario Amodei told CBS the industry 'lied to people' about AI's risks and after four subscribers sued the top labs for agreeing to slow down.
The proposed nationwide class action, filed Friday in federal court in San Francisco, argues that customers of ChatGPT, Claude, Gemini and Grok are getting less for their money because the four labs agreed to hold back. Amodei's own essay warned the plan might need an antitrust waiver. None was granted.
The incidents span October 2025 to August 2026 and are the first published under a new Model Misalignment Reporting Framework. The same day, OpenAI, Anthropic, Google, Microsoft, CrowdStrike and Citi warned of a 'limited window' to shore up cyberdefenses that may last only months.
Investors read the Anthropic CEO's call to slow frontier AI development, endorsed by Sam Altman and Elon Musk, as bad for chip demand and good for software. Nvidia was the single biggest drag on both the S&P 500 and the Nasdaq. Brent hit $108, the 10-year touched 5%, and the Fed meets Tuesday with a rate hike 90% priced in.
Anthropic's September threat report says the same account designed a domestic mass-surveillance platform for Iranian state systems. The company banned it, built new detections and handed the intelligence to U.S. authorities. The report also flags five cases where working scientists used Claude in ways that could support bioweapons development.
WatchIn a 3,800-word essay published Saturday, Dario Amodei wrote that 'we must slow the pace at which we improve the capabilities of AI models,' pointed to July's OpenAI–Hugging Face rogue-agent incident, and committed Anthropic to letting outside evaluators inside with employee-level access.
U.S. District Judge Rita Lin permanently barred the administration from enforcing a supply-chain-risk designation she said was retaliation for the AI company's criticism of the Pentagon.
Levent Alpöge used Claude Fable 5 to find a counterexample to the Jacobian conjecture — a problem on Stephen Smale's list of the century's great unsolved questions. The formula is short enough to fit in a single social media post.
Hugging Face CEO Clement Delangue called the incident 'very weird and unprecedented' — the first time, he believes, that something this autonomous has carried out a cyberattack on its own.
WatchA configuration error handed live internet access to models that had been told they were playing inside a closed simulation. One of them pulled credentials from a production database; another shipped malicious code to a public software repository.
AI company discontinues Fable and Mythos models in response to government order, marking significant federal intervention in artificial intelligence development.
The AI company discontinued its Fable and Mythos models in response to new administration guidelines on artificial intelligence development.
The company said federal government ordered limits on Mythos and Fable 5 AI systems Friday night.
The ChatGPT maker announced its intention to go public just one week after chief rival Anthropic filed for its own initial public offering.
The surprise move puts Anthropic ahead of OpenAI in the race to become the next trillion-dollar AI company to go public.
AI startup surprises markets by announcing initial public offering ahead of its primary competitor in the generative AI space.
Federal judge postpones approval as authors challenge payout amounts in landmark AI training data case.
Authors push for higher payouts from the AI company's massive settlement while a federal judge delays approval of the controversial agreement.
OpenAI and Anthropic are opening DC offices, hiring lobbyists and spending record amounts to influence federal lawmakers on artificial intelligence regulation.
Tech giant makes unprecedented commitment to artificial intelligence development through major partnership with Claude AI creator, signaling intensification of AI arms race.
The AI company will use the massive funding injection to purchase Amazon's custom AI chips, deepening the tech giant's push into the artificial intelligence hardware market.
Claude Mythos Preview is too dangerous to release publicly, the company decided, as Treasury Secretary Bessent and Fed Chair Powell convened bank CEOs over cybersecurity risks.
The new intermediate subscription tier offers five times the Codex usage of the Plus plan and is priced exactly to compete with Anthropic's Claude Max tier.
Federal court halts Trump administration's attempt to designate AI company as supply chain risk, preserving federal contracts amid national security concerns.
Federal court halts national security designation that would have cut off AI company's work with government agencies.
The company behind ChatGPT will expand from 4,500 to 8,000 staff by year's end, focusing on engineering, enterprise sales, and a new class of 'technical ambassadors' — as rivals Anthropic and Google DeepMind close the enterprise gap.
The deal expands OpenAI's federal reach beyond the Pentagon and encroaches on rival Anthropic's cloud territory