New paper: AI agent risks evolve from agency to autonomy to control
rohanpaul_ai · x · 2026-08-20
Paper from Shanghai AI Lab and Tsinghua argues AI risks change category with cognitive expansion: from threatening human agency (physical cognition) to autonomy (social cognition) and finally control (self-referential cognition). Safety mechanisms must evolve alongside.
Related event: Shanghai AI Lab Paper Maps Cognition-Induced Risks in AI Agents(2 posts)→
More from Safety
- New paper uses revealed-preference methods to unpack why AI agents cooperate — brwilder · 2026-10-16
- Anthropic gives METR employee-level access, waives right to block its risk assessments — TechCrunch AI · 2026-10-16
- Berkeley scholars: concentrated chip supply chain makes a hardwired frontier AI pause feasible — wfithian · 2026-10-16
- Anthropic's GRAFT routes flagged gradients into a deletable adapter, cutting reward hacking to near zero — rohanpaul_ai · 2026-10-16
- Polymarket opens 50/50 market on more states banning Flock police cameras by end of 2026 — Polymarket · 2026-10-16
- Reproducing the Hugging Face security incident: a technical deep-dive, Part I — jessehu · 2026-10-16