News for hackers. Stuff that matters.
Re:Claude's New Constitution by viccis on Wednesday January 21, 2026 @10:48PM (#46715075)
By what measure? What's "safe"?
Re:Claude's New Constitution by conception on Thursday January 22, 2026 @08:59AM (#46719319)
https://crfm.stanford.edu/helm/air-bench/latest/#/leaderboar...

This isn’t the gotcha question you think it is. AI safety is being defined and measured.

Re:Claude's New Constitution by viccis on Thursday January 22, 2026 @02:35PM (#46724099)
Cool, another metric to game like they do the other ones.