Category Anthropic

Developments in AI safety and latest research from Anthropic.

Cursor Brings Its Coding Agents to iPhone as AI Programming Goes Mobile

Cursor’s new iPhone app lets developers manage a coding agent on the go, highlighting the shift toward mobile AI programming.

California strikes a Claude deal with Anthropic, giving state and local agencies discounted AI access as Newsom pushes AI in government.

Z.ai says GLM-5.2 boosts cybersecurity AI, with researchers noting strong bug-finding skills and new policy concerns.

Anthropic’s AI policy standoff with the Trump administration is exposing major risks for the company and the U.S. AI industry.

OpenAI chip plans with Broadcom show Big Tech is moving to custom silicon to cut costs and reduce reliance on Nvidia.

AI model approvals are becoming a key bottleneck as the U.S. government tightens scrutiny of frontier releases from OpenAI and Anthropic.

Europe’s AI sovereignty drive is accelerating as Trump-era policy shocks expose the risks of relying on American models and infrastructure.

Mythos 5 is back for approved users after talks with the Trump administration, but Anthropic’s Fable 5 still faces unresolved restrictions.

The Trump administration restores Anthropic Mythos 5 access for select U.S. agencies and companies after a cybersecurity-related ban.

Claude Mythos 5 gets partial access back after a U.S. review, signaling tighter government control over frontier AI releases.