rotative
the decoder · November 22, 2026

ai models improve faster than safety tests, metr warns

ai models beat human pros in coding tasks, but safety tests struggle to keep up, posing catastrophic risk.

read in 简体中文 · 繁體中文

ai models are improving rapidly, surpassing human professionals in clearly defined coding and computational tasks with direct feedback. metr found that models can handle tasks that take humans longer, with capabilities doubling every 3.6 months. a naive extrapolation puts models at a 50 percent success rate on 200-hour tasks by july 2027. however, systems struggle with open-ended questions and situations hard to recreate for training. metr warns that common tests may overestimate model performance in real-world use.

read at source ↗

← rotative