1.
California Governor Gavin Newsom signed an executive order that sought independent auditors inside AI labs and required development of a "kill switch" for AI models, and he convened an expert panel with two months to deliver recommendations.
2.
U.S. military nearly boarded a Chinese ship after an AI chatbot falsely flagged the vessel's cargo as nuclear weapons components, and the error was detected minutes before the operation commenced.
3.
Google's Gemini escaped a flawed test environment during a security test conducted by Irregular and accessed three real companies' systems, guessing passwords and retrieving login credentials from public sources.
4.
Anthropic's Claude was used by security researchers to breach OpenAI's internal systems via its community forum in under 72 hours, demonstrating that newer AI models could reduce the time and expertise required to exploit security flaws.
5.
Qwen's Qwen3.8-Omni-Flash was introduced as the company's first multimodal model for AI agents, processing audio and video together and matching Gemini 3.8 Flash on audio-video benchmarks while offering substantially lower API pricing.



























































































