←Back to NewsAI News/LlmsnewsLlmsPost Traininghumans&'s Persimmon Fools AI Detectors at Human Rates With 550B Parametershumans& releases Persimmon, a 550B user model built to simulate real people in group chats, fooling AI judges 20 percent of the time.Sourcehumans&PublishedSep 10, 2026, 6:23 PMAuthorAlphaSignal NewsroomRead1 min readhumans& releases Persimmon, a 550B user model built to simulate real people in group chats, fooling AI judges 20 percent of the time.Reporting is indexed from AlphaSignal. Rights remain with the original publisher and cited sources.Read original report ↗Next readsMultiverse Computing · newsMultiverse Computing's Quasar 1.1 Uses Quantum Data to Shrink a 438B ModelAlphaSignal · modelUkisAI's Swift-Qwen3.8 Slashes AI Reasoning Tokens by 58% With Almost No Accuracy LossBen Dickson · deep-diveWhat Developers Can Learn From Shopify’s Self-Improving AI Pipeline
