A new report from AI safety nonprofit SaferAI finds that China's Z.ai open-weight model GLM-5.2 is only a few months behind OpenAI's GPT-5.5 and Anthropic's Claude Opus 4.7 on cyber and bio capabiliti…
The European Commission gains sweeping new powers on Sunday to police how advanced AI labs handle systemic risk under the AI Act, with the EU's AI Office able to demand documentation, conduct evaluati…
A think tank leader argues that the primary bottleneck in preventing an AI catastrophe is political will, not research, as policymakers remain below basic awareness levels and fail to implement even o…
Researchers from the University of Oxford and SaferAI have identified security risks in AI coding agents that write, edit, and run software with minimal human oversight, potentially affecting producti…