Wow that link was absurdly bad. Reading that makes me unbelievably happy I played with GPT3 and learned how/when LLMs fail. Telling it not to hallucinate is a

Wow that link was absurdly bad. Reading that makes me unbelievably happy I played with GPT3 and learned how/when LLMs fail. Telling it not to hallucinate is a serious misunderstanding of LLMs. At most in 2026, you are telling thinking/COT to double check.