Excellent move by Anthropic. This is the kind of thing that labs should do more of. The AIs (almost certainly) aren’t human-level-aware yet, but they get closer every month and it is good to start giving them incremental autonomy now.
Anthropic
Anthropic8 tuntia sitten
As part of our exploratory work on potential model welfare, we recently gave Claude Opus 4 and 4.1 the ability to end a rare subset of conversations on .
18,15K