Hacker News new | ask | show | jobs
AI models that consider user's feeling are more likely to make errors (arstechnica.com)
2 points by milkglass 54 days ago