Hacker News
- ChatGPT bombs test on diagnosing kids' medical cases with 83% error rate https://arstechnica.com/science/2024/01/dont-use-chatgpt-to-diagnose-your-kids-illness-study-finds-83-error-rate/ 7 comments
- ChatGPT bombs test on diagnosing kids’ medical cases with 83% error rate | It was bad at recognizing relationships and needs selective training, researchers say. https://arstechnica.com/science/2024/01/dont-use-chatgpt-to-diagnose-your-kids-illness-study-finds-83-error-rate/ 94 comments technology
Linking pages
- Large Language Models Are Drunk at the Wheel | Matt.si https://matt.si/2024-02/llms-overpromised/ 366 comments
- Cobalt Core is a tight, funny roguelike deck-builder deserving of many runs | Ars Technica https://arstechnica.com/gaming/2024/01/cobalt-core-is-your-next-just-one-more-roguelike-deck-building-habit/ 1 comment
- BMW showed off hallucination-free AI at CES 2024 | Ars Technica https://arstechnica.com/cars/2024/01/bmws-ai-powered-voice-assistant-at-ces-2024-sticks-to-the-facts/ 0 comments
Linked pages
- Cobalt Core is a tight, funny roguelike deck-builder deserving of many runs | Ars Technica https://arstechnica.com/gaming/2024/01/cobalt-core-is-your-next-just-one-more-roguelike-deck-building-habit/ 1 comment
- High-performance medicine: the convergence of human and artificial intelligence | Nature Medicine https://www.nature.com/articles/s41591-018-0300-7 0 comments
Related searches:
Search whole site: site:arstechnica.com
Search title: ChatGPT bombs test on diagnosing kids’ medical cases with 83% error rate | Ars Technica
See how to search.