SAFE

A new arXiv paper (2609.17865v1) introduces SAFE, a controlled benchmark testing whether frontier models acquire safety-relevant evidence before making deployment decisions. Across GPT-5.5, o3, Claude…

A commentator argues that AI models should have built-in controls to refuse harmful requests, comparing the need for friction to restrictions on handguns and fentanyl. The author advocates for open-so…

European Central Bank research finds that jobs vulnerable to AI substitution declined in the US between 2019 and 2025, while firms adopting AI are 4% more likely to hire and see 4% labor productivity …