Study: ChatGPT dating advice beats humans blind, but AI label flips ratings
ipeirotis · x · 2026-09-10
A preregistered study by Osborne & Bailey (Scientific Reports, 2025; 5 experiments, N=1,722) finds that blind raters judge ChatGPT's advice as higher quality, more effective, and more authentic than human advice — even in dating/relationships, where people strongly prefer human advice. Provenance labels flip judgments: the same poems rated worse once labeled AI-generated, even though AI poems actually scored better. People's aversion is provenance bias, not quality.
Related event: ChatGPT Beats Humans in Blind Evaluation of Relationship Advice(2 posts)→
More from Research
- Goodfire commits $10M to launch a 'Human Genome Project for AI' interpretability effort — mathildepapillo · 2026-10-27
- Researchers call for a Human Genome Project for AI to fully understand neural networks — mathildepapillo · 2026-10-27
- Jon Barron: every softmax and cross-entropy implementation has two subtle numerical bugs — jon_barron · 2026-10-27
- New theory paper explains what Anthropic's Jacobian lens actually detects in LLMs — wgilpin0 · 2026-10-27
- Interpretability researcher joining Goodfire AI post-PhD, calls for speeding up the field — mathildepapillo · 2026-10-27
- EdgeRock-4B holds KV cache at 72 MiB across 256K context, 99.1% smaller than Qwen3.5-4B — alifcoder · 2026-10-27