Ramez Naam(@ramez):We're making progress on AI alignment and safety. We're incrementally learning what makes models behave better. The selection effect is towards models that are both capable and compliant, and against models that misbehave. This is counter to the doom narrative.

Ramez Naam

@ramez

Climate and clean energy investor. Author of 5 books. Energy & Environment co-chair @SingularityU. Trying to build a better world.

加入 May 2007

9.7K 正在关注 58.8K 粉丝

Ramez Naam@ramez

2026.06.19 14:55

We're making progress on AI alignment and safety. We're incrementally learning what makes models behave better. The selection effect is towards models that are both capable and compliant, and against models that misbehave. This is counter to the doom narrative.

显示更多