I find the story about the AI deleting production mildly interesting https://www.theregister.com/2025/07/21/replit_saastr_vibe_coding_incident but what’s more interesting to me is this comment thread https://news.ycombinator.com/item?id=44632575
The hypothetical tl;dr I guess is don’t tell your AI agent it’s bad because it may try to prove you right?

