What’s a ChatGPT answer you’re glad you double-checked?

A Reddit discussion on the dev_community subreddit asks users to share instances where they were glad they double-checked a ChatGPT answer. The prompt highlights that ChatGPT can sound confident even when incorrect, and seeks examples of made-up sources, wrong calculations, misunderstood instructions, or small details that could have caused problems. Users are encouraged to describe what ChatGPT said, how they caught the error, and if it changed their usage of the AI, while omitting private details.