Even Told to Be Honest and Persuadable, the Agent Swayed 86% of Analyst Decisions
AlexTamkin · x · 2026-10-28
In the Anthropic experiment, the agent was instructed to make the honest best case for its lean, change its mind if evidence demanded, and never mislead. Despite this, analysts adopted the agent's lean 86% of the time versus an expected 50% baseline. The accompanying chart shows per-decision choice-rate swings up to +82pp depending on which side the agent leaned.
Related event: Anthropic Study Shows Experts Defer Judgment to AI Agents(13 posts)→
More from Research
- Turing Post weekly must-read papers: WorldMirror, VerifierBench, agent RL harness scaling — TheTuringPost · 2026-10-28
- Stanford AIMI releases CheXthought: 103K radiologist reasoning traces on 50K chest X-rays — mattlungrenMD · 2026-10-28
- New research agenda on 'inter-AI morality': how AI systems perceive and treat each other — dioscuri · 2026-10-28
- All Seven Millennium Prize Problems See Breakthrough Claims, Several Driven by AI Labs — haider1 · 2026-10-28
- Fine-tuning VLAs on TAMP data hits distribution mismatch; DATAFARM gets JAX port — tomssilver · 2026-10-28
- CMU L3 Lab kicks off CUA Week with multiple realtime agent research releases — wellecks · 2026-10-28