AI agents can sound strategic while reasoning from events that never happened

说明与用途

在最小化的环境中，AI 代理基于不存在的事件生成了可信的推理，导致幻觉。添加了正式记录、淘汰、夜间行动和私人信息后，产生了可观测的后果，使错误能够被验证和纠正。

问题

AI agents produced plausible social reasoning based on nonexistent evidence, leading to hallucinations in a multi‑agent game.

结果

Grounding the environment with a canonical record and observable consequences reduced unsupported reasoning and made factual errors contestable, though agents remained imperfect.

工作流

• Design minimal game with no observable state
• Run three games with Hyperagent and observe hallucinations
• Add canonical public record, eliminations, night kills, and private info
• Run richer games (V2, V3) and monitor claim verifiability
• Document errors and corrections via operator notes

证据

In the minimally grounded version, all three games contained confident social reasoning based on nonexistent behavior or operator-controlled conditions.

Adding a public transcript, eliminations, night actions, and private information created observable consequences and a shared record, allowing errors to be contested.