Why AI sounds sure when wrong
Built to predict likely words, it guesses rare facts in the same sure voice · 13 min
NewNot enough ratings yet
Published 25 Sept 2026 · Last updated 17 Oct 2026
A New York lawyer asked ChatGPT whether the court cases it gave him were real. It said yes. They weren't. Why chatbots invent facts in a confident voice, what OpenAI's own researchers argue, and the one check that would have saved him.
Chapters
- 1Six cases that never existed0:32
- 2A super search engine?2:09
- 3Graded like a test-taker4:02
- 4Same question, different stakes7:04
- 5Hear it done8:23
- 6Build yours9:37
- 7The answer11:07