A recent experience with ChatGPT 5.5 Pro | Hacker News

A Hacker News post described using ChatGPT 5.5 Pro in a fresh session to critique difficult proofs and stressed that AI output should be independently verified.

The post warned that unreviewed AI work can waste experts’ time, create unnecessary cleanup, and fuel growing “AI slop” fatigue.

The broader takeaway is that stronger AI makes human review even more important before code, proofs, or findings are shared or shipped.