2 weeks ago
Fri Jan 23, 2026 4:01pm PST
AI hallucinate. Do you ever double check the output?
Been building AI workflows and then randomly hallucinate and do something stupid so I end up manually checking everything anyway to approve the AI generated content (messages, emails, invoices,ecc.), which defeats the whole point.

Anyone else? How did you manage it?

comments:
add comment
loading comments...