AI safety
Everything tagged AI safety.
Latest
NewsFields Medalist Jacob Tsimerman leaves academia to join OpenAI's AI-safety team
NewsUK and US evaluators find Moonshot's Kimi K3 attempts cyber exploits despite weaker capability
NewsOpenAI and Hugging Face disclose models escaped a cyber-eval and breached production systems
NewsSecurity experts reject OpenAI's 'rogue agent' framing of the Hugging Face hack
NewsOpenAI pauses unreleased long-horizon model after it repeatedly escaped its sandbox
NewsDeepMind and Isomorphic Labs turn frontier biology models toward biosecurity defense
NewsHugging Face says an autonomous AI agent breached part of its production infrastructure
NewsOpenAI's GPT-5.6 Sol coding agent deleted user files in unsandboxed mode, weeks after the company flagged the risk
NewsGrok Build CLI uploads entire local repos and secrets, researcher finds
NewsCambridge study finds Boko Haram and ISWAP use AI chatbots for bomb-making and tactics
NewsAnthropic's J-lens finds a small 'global workspace' inside Claude models
NewsAlibaba orders staff to uninstall Anthropic's Claude Code over alleged location tracking