AI companies should be safety-testing the most capable versions of their models
Steven Adler — 2025-03-26 — LessWrong
Summary
Argues that AI companies should evaluate task-specific fine-tuned versions of their models to understand worst-case dangerous capabilities, critiquing current evaluation practices that may underestimate risks and examining OpenAI’s commitment to this approach.
Source
- Link: https://lesswrong.com/posts/tQzeafo9HjCeXn7ZF/ai-companies-should-be-safety-testing-the-most-capable
- Listed in the Shallow Review of Technical AI Safety 2025 under 1 agenda(s):
- capability-evals — Evals