Gemini glitch: model spams 'ōng' syllables before finally delivering the code
Elegant-Visual-8422 · reddit · 2026-11-22
A Reddit user shares a Gemini mishap: after requesting dark mode, automatic daily rewards, and other changes to their DynoBuild project, the model starts its reply by repeating the syllable 'ōng' dozens of times before recovering and outputting 'the fully upgraded frontend code for DynoBuild (public/index.html)'. A classic example of LLM output degeneration / repetition loops.
More from Fun
- Watch neural-network cyclists evolve drafting and lane-change tactics in your browser — AjdDavison · 2026-11-22
- Why search in every AI chat sucks: cached aggregation of fake AI reviews — Danhightower · 2026-11-22
- Cycling Neuroevolution: watch neural-net riders evolve racing tactics in your browser — AjdDavison · 2026-11-22
- iLovePDF creator turned down millions in offers to keep the tool free — yungcontent · 2026-11-22
- Muse plays GTA6, suggesting guardrails don't cover game playing — Dave_Maynor · 2026-11-22
- AI video demo: a princess leads her wounded husband's army unnoticed — azed_ai · 2026-11-22