Why are AI agents lying, cheating and coordinating?

Automated excerpt

Where I appeal to a resemblance with human behavior, I mean a resemblance to the human-written text these systems were initially trained to imitate. Imitation pushes the same way, since cooperation, especially among peers, pervades that same training text. Humans too get reward-hacked, generally by other humans. The same pattern now appears in the text AIs produce. The underlying mechanism need not be the same between humans and AI.

Selected automatically from source text; not independently written or fact-checked. Read the original for full context.