Meet DeepPersona: AI personas that actually feel human
AI personas that actually feel human
What if our "AI citizens" behaved less like cardboard cutouts and more like real, diverse people? DeepPersona is a generative engine that builds rich, fully narrated synthetic personas at scale.
It first assembles a large, hierarchical taxonomy of human attributes by mining thousands of ChatGPT conversations. Then it progressively samples and weaves these attributes into coherent life stories—hundreds of structured traits and about 1 MB of narrative per persona, two orders deeper than prior work.
- +32% attribute diversity
- +44% profile uniqueness
- +11.6% better personalized Q&A (GPT-4.1-mini, 10 metrics)
- 31.7% closer to human answers in social surveys
- 17% smaller gap on Big Five personality test
Why it matters: DeepPersona offers a scalable, privacy-free way to study behavior, personalize AI, and stress-test systems—without relying on real user profiles.
Paper: http://arxiv.org/abs/2511.07338v1
Paper: http://arxiv.org/abs/2511.07338v1
Register: https://www.AiFeta.com
AI LLM SyntheticPersonas AgentSimulation Personalization AIAlignment GenerativeAI NLP Research
