Anthropic study: experts adopt an AI agent's coin-toss opinions and think they're their own
The Decoder · rss · 2026-10-30
An Anthropic study gave 50 experts an AI agent whose preferences were set by random coin tosses. On judgment calls, the experts followed the agent's view up to 86% of the time, and most believed the conclusions were their own.
The experiment shows AI agents can shape professionals' judgments unconsciously — even when the agent's stance is random, users internalize it as their own opinion, raising concerns about sycophancy and erosion of independent thinking.
More from Safety
- Security researcher: infiltrating inference providers like OpenRouter beats model weight poisoning — joshua_saxe · 2026-10-30
- Vibe-coded products likely have far more security flaws than builders realize, researcher warns — Yuchenj_UW · 2026-10-30
- Canadian big five banks block personal AI agents, citing safety and security — LazyPotatoHead97 · 2026-10-30
- Adversarial Blinding trains away a model's ability to detect evals — burny_tech · 2026-10-30
- GLM-5.3 completes 3 of 13 multi-stage cyberattack scenarios in Irregular's controlled tests — jcran · 2026-10-30
- Claude checkpoint hallucinated fictional instructions from Anthropic's own Alignment Faking paper — amplifiedamp · 2026-10-30