FoodSafetyAI joke model goes viral: whatever you ask, it answers "I would toss it"
chrisalbon · x · 2026-10-30
Chris Albon posted a parody open-source model launch, FoodSafetyAI, claiming it was distilled from OpenAI Astra, "100% effective at preventing foodborne illnesses," scalable on small mobile devices and passing the toughest AI safety benchmarks — with a trash-can logo as the reveal. The entire "model" is one line of code: def foodsafetyai(userprompt): return "I would toss it." No matter the input, it says "I would toss it." A sharp satire of AI launch-speak (distillation, open source, safety benchmarks, 100% efficacy claims) that has spread widely in the AI community.
More from Fun
- AO3's 2-4 week downtime sparks speculation: rewriting everything with AI? — jd_pressman · 2026-10-30
- Weirdest GPT-pro output yet: model leaks incoherent inner monologue into the reply — jessi_cata · 2026-10-30
- Short film with an all-AI cast is indistinguishable from real actors, says viewer — TotGaH · 2026-10-30
- Arizona asks residents to limit showers to 30 seconds while AI data centers drink its water — CtrlAltDwayne · 2026-10-30
- Everyone is trying to look effortlessly unique in exactly the same way — signulll · 2026-10-30
- Hailuo AI shares Halloween-themed mascot art in four costumes — Hailuo_AI · 2026-10-30