It does sometimes go a little bit nuts, and OpenAI continues to be sincere with regards to the 'hallucinations' that ChatGPT might have, and the issues inherent in these LLMs. Be cautious of relying entirely on everything ChatGPT states.It'd be doable to struggle a website traffic ticket with an AI 'robotic attorney' secretly feeding you traces in