SnitchBench – AI Model Whistleblowing

Posted by charlieirish on Monday June 02, 2025 @10:49AM
from the snitchbench.t3.gg dept.

( 1 comment | on HN )

Re:SnitchBench – AI Model Whistleblowing by tamentis on Tuesday June 03, 2025 @02:23AM (#44166959)

If it goes through the wire, don't trust it.

[ Parent | on HN ]