White House blocks AI models from UK testers
The White House told OpenAI and Anthropic to stop sharing new models with Britains AI Safety Institute until U.S. officials finish their own review.
2026-09-2511 views
The White House told OpenAI and Anthropic to stop sharing new models with Britains AI Safety Institute until U.S. officials finish their own review.
OpenAI and Anthropic CEOs told the UN Security Council that AI could threaten humanity — while the US and China clashed over governance.
Ten days after Dario Amodei urged the industry to pace AI development, Anthropic released Claude Opus 5.5 — beating GPT-6 Astra on six of eight benchmarks at 40% lower cost.
Four consumers filed a class-action lawsuit accusing Anthropic, OpenAI, SpaceXAI, and Google of Sherman Act violations for coordinating to slow AI development.
Anthropic partners with Accenture on embedded evaluation, each investing $1B over 5 years. Third parties get employee-level access to training pipelines.
Anthropic publishes the first hard numbers on recursive self-improvement — and the safety spending looks thin.
OpenAI surpasses Anthropic in developer spending for the first time since 2024, as GPT-6 Sol appears in secret testing — six times faster than Astra.
Three frontier AI labs have been negotiating safety standards privately for weeks, behind the public slowdown campaign.