Study: mutual in-context learning between AIs amplifies noise and accelerates collective belief collapse
Hidenori8Tanaka · x · 2026-09-09
Hidenori Tanaka presents a thread on the team's Quantized Simplex Gossip (QSG) model, which represents AI output as sampling symbols from internal belief distributions, with persona plasticity α controlling the listener's adaptation.
Key mechanism: shorter messages contain fewer sampled symbols, giving a noisier picture of the speaker's beliefs. The theory predicts that mutual in-context learning among AIs can amplify these fluctuations and accelerate collective belief collapse — a theoretical framework for information degradation in multi-agent communication.
More from Models
- 'Doomer takeover = AI winter': Terra 6.1 flops on Terminal-Bench 4.0 vs Sonnet 5.5 — mark_k · 2026-10-27
- Together AI sits down with Meta to explain Muse Spark Open — togethercompute · 2026-10-27
- GPT-6.1 Terra hits ~60% on Agents' Last Exam at $1/task, 9x cheaper than Opus 5.5 — ChrisGPT · 2026-10-27
- ChatGPT keeps telling tech-support users to email 404 Media's Jason Koebler — petrusenko_max · 2026-10-27
- GPT-6.1 Terra called the awkward middle tier, priced at half of GPT-6.1 Sol — cedric_chee · 2026-10-27
- Mystery model "Terra" scores 48 on Artificial Analysis index with no public API — haider1 · 2026-10-27