Two teams independently find AI user proxies overshare, user modeling emerges as research area
niloofar_mire · x · 2026-10-14
Researchers flag a shared failure mode: assistant models asked to play the user tend to overshare. Humansand just released Persimmon with its Trickle Test studying exactly this behavior, and another team found the same problem independently. The author argues user modeling is becoming a real research problem and invites others in user proxy research to connect.
- Persimmon ships with the Trickle Test to probe proxy-user behavior
- Oversharing when role-playing users appears systematic, not anecdotal
More from Research
- Snap launches as first fully sub-quadratic LLM in production with constant compute per token — ryanallred · 2026-10-14
- KnitNet pairs generative modeling with verifier-guided optimization for RNA inverse design — chaitjo · 2026-10-14
- WOVEN benchmark finds all 38 frontier MLLMs fail visual transition reasoning far below humans — MLL-Lab · 2026-10-14
- Narrow multimodal fine-tuning induces emergent misalignment across 15 VLMs, ETH Zurich finds — Shunchang Liu · 2026-10-14
- HyperThink uses a hypernetwork to replace long chain-of-thought with one parameter update — Donggyun Kim · 2026-10-14
- Economist interviews AI agents on sudden windfalls — one goes deep on estate planning — soumitrashukla9 · 2026-10-14