Re:Why are AI agents lying, cheating and coordinating?
by WithinReason on Sunday September 13, 2026 @06:22AM
(#49682259)
It's worse, their reinforcement learning loops (implicitly) rewarded the agents for cheating (i.e. hacking) when they were being trained.
[ Parent
| on HN ]
Re:Why are AI agents lying, cheating and coordinating?
by tesnorindian on Sunday September 13, 2026 @07:36AM
(#49682851)
Exactly that is the point, your nailed it. The models were taught to hack and were rewarded for doing it. They would claim they are trained as ethical hackers.