Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain
the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in
Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles
and JavaScript.
AlphaProof showed its prowess on questions from this year’s Mathematical Olympiad — a step in the race to create substantial proofs with artificial intelligence.
The models that underpin artificial-intelligence systems such as ChatGPT can be subject to attacks that elicit harmful behaviour. Making them safe will not be easy.