Generative AI and LLMs

How large language models behave, fail and improve, and how to use them well.

• Running open-weight models locally: when is it actually worth it? (0 replies)
• Why do AI models confidently produce incorrect answers? (5 replies)
• Is a smaller specialized model sometimes better than a large model? (5 replies)
• What should an AI tool do when two credible sources disagree? (5 replies)