2 stories in this blend

During an event marking Anthropic's transition to a publicly traded business, chief executive Dario Amodei committed to halting model rollouts whenever safety checks remain incomplete. He argued that setting an example of disciplined caution would force other developers to match rigorous control standards rather than accelerating purely for market gains.

Executives from major tech companies, including Google, OpenAI, Meta, Anthropic, Nvidia, and xAI, agreed to a new non-binding safety framework. Federal agencies were also instructed to use designated terminology when referring to advanced artificial intelligence.