Bedrijf

February 3, 2026, 5:16 AM EST. An annual snapshot, the International AI Safety report charts progress and risks from deepfakes to the jobs market. Chaired by Yoshua Bengio and guided by Nobel laureates Geoffrey Hinton and Daron Acemoglu, the document is described as a state-of-play piece, not a policy blueprint, but expected to frame talks at the upcoming AI summit in India. Key takeaways: 1) AI models are improving. New systems such as OpenAI's GPT-5, Anthropic's Claude Opus 4.5 and Google's Gemini 3 show stronger reasoning and faster gains in maths, coding and science, though they remain prone to hallucinations and cannot execute long, complex tasks autonomously. 2) Deepfakes are proliferating; the report notes harder-to-distinguish AI content and cites 15% of UK adults seeing AI-generated imagery and 77% mislabeling ChatGPT text as human. 3) AI firms face emerging biosecurity and chemical risk considerations.