Anthropic: Emergent Misalignment from Reward Hacking (anthropic.com)

1 points by highfrequency 1mo ago ↗ HN

0 comments

[ 0.28 ms ] story [ 6.3 ms ] thread

No comments yet.

A showcase of shard-db. Data from the Hacker News API. Refreshed every 15 minutes.

Source · Live DB stats