GPTZero launches 4o detector with 1/10,402 false positive rate, robust to paraphrasing attacks
sethlazar · x · 2026-09-25
AI detection startup GPTZero has launched GPTZero 4o, an all-purpose AI text detection model it says achieves a false positive rate of 1/10,402 on benchmarks and is its most robust model yet against edits and paraphrasing attacks used to evade detection. Cornell professor Seth Lazar welcomed the competition, saying academic editing needs all the help it can get against the advancing tide of AI-generated slop.
More from Models
- Slept through Codex auto top-up draining his card — bank blocked the charges as fraud — CtrlAltDwayne · 2026-09-25
- Tom Dietterich: SFT and RL go beyond next-token prediction in LLMs — tdietterich · 2026-09-25
- Community Speculates DeepSeek V4 Coming Soon After Holiday Post From Liang Wenfeng — teortaxesTex · 2026-09-25
- NaceAI launches Drex, a sub-6B decision model that tops the public Decision Index at 51.73 — ordax · 2026-09-25
- Model audit showdown: Astra dominates, Opus and Fable close, Grok 4.7 and GPT-6 Sol lag far behind — ivan_bezdomny · 2026-09-25
- Uncensored local model Bonzai 2 27B tops benchmarks, runs on 12GB VRAM — alexcovo_eth · 2026-09-25