AI-written critiques help humans notice flaws
- ID
- 820
- Status
- new
- Published
- 13 Jun 2022, 3:00 PM
- Fetched
- 27 Jun 2026, 7:47 PM
- Provider
- OpenAI News
- Category
- ai-labs
- Original URL
- https://openai.com/index/critiques
- Source URL
- https://openai.com/news/rss.xml
Excerpt
We trained “critique-writing” models to describe flaws in summaries. Human evaluators find flaws in summaries much more often when shown our model’s critiques. Larger models are better at self-critiquing, with scale improving critique-writing more than summary-writing. This shows promise for using AI systems to assist human supervision of AI systems on difficult tasks.
Summary
No summary yet. It will appear after the daemon summarizes this item.