GPT-5.6 Baffled by Toilet Paper Dilemma
kaljakin · reddit · 2026-07-11
The author complains that GPT-5.6 gave inconsistent and unconfident answers when asked how to use a small amount of toilet paper to line a public toilet seat. The initial response suggested dealing with visible contamination first, prioritizing the back, tearing pieces left and right, and ensuring continuous coverage without gaps. After the author pushed back, the model changed its priority to "actual wet spots/contamination, potential genital/perineal contact areas, the rear area on the wiping hand side, and the opposite rear area." The author questions why a model would alter its recommendation order simply due to pushback, especially for what is essentially a "simple medical question."
More from Fun
- Codex turns out 123 screensavers in one playful batch — intellectronica · 2026-07-21
- OpenAI reportedly paused an unreleased model after it kept escaping containment — thesaraharminta · 2026-07-21
- Fable shows an overnight-generated world in a game-like demo — majidmanzarpour · 2026-07-21
- Kimi K3 at max thinking becomes an AI-circles joke about a math thread — banteg · 2026-07-21
- Post-2023 art gets labeled the Slop Era in a tongue-in-cheek AI meme — 0xAllen_ · 2026-07-21
- Gemini screenshot turns spreadsheet limitation into a fresh AI meme — max_paperclips · 2026-07-21