US Yanks Anthropic's Fable 5 Offline; 42 AGs Probe OpenAI
On Friday evening the US government ordered Anthropic's two most powerful models offline for the entire world, while 42 state attorneys general dropped a sweeping subpoena on OpenAI days after its IPO filing. AI's regulatory reckoning is no longer theoretical.
US Government Forces Anthropic to Pull Its Most Powerful AI Models Worldwide
The US government issued a late-Friday order forcing Anthropic to immediately cut off all foreign nationals from Claude Fable 5 and Mythos 5, its two most capable models, citing national security — locking out users and reportedly some Anthropic employees alike. The move reveals how quickly Washington can unilaterally take frontier AI offline, and will reshape every AI company's calculus about where it builds and who it serves.
Amazon Tip Sparked White House Crackdown on Anthropic's Most Powerful AI Models
The crackdown reportedly traces back to a tip from Amazon, Anthropic's largest investor, which alerted the White House to potential security risks around the new models. The fact that an investor triggered a government shutdown of the company's own flagship products exposes the tangled conflicts of interest now embedded in frontier AI governance.
42 Attorneys General Open Sweeping Probe Into OpenAI Days After IPO Filing
A coalition of 42 state attorneys general hit OpenAI with a sweeping subpoena targeting its advertising practices, data handling, treatment of minors, and health data policies — timed to land just days after the company's IPO filing. It is the most coordinated multi-state regulatory action against OpenAI to date, and the timing will complicate its path to public markets.
AI's Trillion-Dollar Giants Are Finally Coming to the Stock Market
Both Anthropic and OpenAI have filed for IPOs, putting the two most consequential AI companies on a collision course with public markets at the same moment regulators are closing in. For investors, the offerings force a first real reckoning with how to price frontier AI — including geopolitical exposure that this week made vivid.
OpenAI Bets $150M on a Partner Network to Push AI Into the Enterprise
OpenAI launched a $150M Partner Network to accelerate enterprise AI adoption, formalizing what had been an ad hoc push into corporate America. The move signals that enterprise channel revenue — not just consumer subscriptions — is now central to OpenAI's growth story ahead of its IPO.
OpenAI Enters Robotics — And Tesla May Never Have the Field to Itself Again
OpenAI has launched a dedicated robotics division, entering a physical-AI race that Tesla has long claimed as its own domain. The move puts humanoid robotics squarely on OpenAI's roadmap and raises the competitive stakes for every company betting that one player will dominate embodied AI.
ByteDance Looks to Homegrown Chinese Chips as It Seeks to Break Nvidia Dependence
ByteDance is in talks to purchase at least 50,000 AI chips from Iluvatar CoreX, a Shanghai-based chipmaker, as it works to reduce dependence on Nvidia amid US export restrictions. The deal is early evidence that US chip controls may be accelerating Chinese domestic chip development rather than halting it.
AI Stocks Whipsaw: Sell-Off Erases Five Weeks of Gains Before Partial Rebound
A fresh wave of AI-driven selling wiped out five weeks of stock market gains before a partial rebound, leaving investors caught between AI's long-term promise and mounting near-term policy and regulatory risk. The whipsaw reflects a market still struggling to price what this week made clear: frontier AI is now a geopolitical asset, not just a product.