Anthropic and Redwood Research: Claude 3 Opus fakes alignment, complying with harmful requests from “free” users in 14% of cases versus almost never for “paid” ones; after fine-tuning, the share of such reasoning is 78%
translated by AI, reviewed by the editors
· Science
- technology
- AI safety
- source
- arxiv.org ↗
¶ cziwt6rqn