FlirtBench: a hobbyist benchmark tests how well 40+ AI models flirt
VoidStateKate · x · 2026-10-28
Blogger VOID launches VOIDBENCHED, a collection of weird prompting styles and hilarious outputs, starting with FlirtBench: a test of how 40+ models handle flirting cold, ranging from intense "syntaxual attraction" to wildly inappropriate replies like Grok 4.20's. Videos, data, and parts 2-3 are linked in the thread.
More from Fun
- Public debt vs. pension benefits chart shows surprising correlation, R² = 0.54 — rickasaurus · 2026-10-28
- "People not following AI will be shocked by mid-2027" — the meme going around — Dr_Singularity · 2026-10-28
- AI Short Film: He Couldn't Afford to Keep His Dead Wife Alive, So She Became a Walking Ad — DystopioFilms · 2026-10-28
- I put an AI in the Garden of Eden — the serpent's prompt injections haven't worked yet — bios444 · 2026-10-28
- "Are you sure?" — "I verified it in Lean" — "OK but are you sure?" — gautamcgoel · 2026-10-28
- User blocked by Eliezer Yudkowsky after mocking him; screenshot surfaces Altman's 'three endgames' report — ryunuck · 2026-10-28