Asking ChatGPT about cube-shaped wombat poo triggers a biosecurity safety flag
myrandomtips · reddit · 2026-10-12
A Reddit user asked ChatGPT voice mode to research why wombats produce cube-shaped poo and write a Google Doc — and the conversation got flagged with "multiple flags for possible biological safety risk," putting responses under extra biosecurity checks. The user sat through roughly 20 minutes of "give me a moment" before seeing the warning on screen. They understand safeguards around genuine bio risks but wonder what word combination tripped it, and argue voice mode should tell users when a safety system is holding up an agent instead of stalling silently.
More from Fun
- AI-generated "lost" E.T. cut scene shows how convincing synthetic film clips have become — bennash · 2026-10-13
- Dev writes custom AI-powered firmware for his MagSafe cable because AI lets him do anything — banteg · 2026-10-13
- 19-year-old Ethiopian developer leaves home with one bag to build his startup in Singapore — filiksyos · 2026-10-13
- From wiring circuits to writing prompts: the 70-year 'degradation' of programmers as a meme — JonathanRoss321 · 2026-10-13
- Creator unveils Gubbu, first AI virtual influencer in a planned character crew — CurieuxExplorer · 2026-10-13
- Everyone Says Muse Wonder Caught the Frontier, but Nobody Actually Tried It — andrey_kurenkov · 2026-10-13