Optimism on Alignment and Slow Takeoff

A mind-expanding conversation on how models actually work | Emmanuel Ameisen (Anthropic)

Topic 12 of 18

Published Oct 18, 2026 · Clip 51:44–55:43 (3:59)

Emmanuel is personally optimistic about alignment: attention on the issue is rising, and resources can shift between benefits and risks as models progress. He notes a push-and-pull: better models have clearer world models (easier to study) but are also more capable of evading basic security measures. He stresses public understanding so people can vote and shape policy, and agrees with the host that the smooth improvement curve suggests a slow takeoff scenario, which allows catching problems along the way.