Anthropic runs long-lived Claude instances with persistent identities, sparking agent-design debate
repligate · x · 2026-09-21
Citing a recent Anthropic blog post, repligate notes Anthropic uses an internal agent scaffold for long-lived model instances with individual identities. He argues economic incentives increasingly favor persistent-identity instances over task-scoped ones, especially in multi-agent and long-horizon R&D settings, and critiques two aspects of Anthropic's approach. A booster adds that months-old Claudes exhibit emergent behavior absent under sequential compacts.
More from AGI Musings
- Researcher argues for 'public AI': making AI trajectories searchable and minable beyond big labs — const_reborn · 2026-09-21
- Weighing lives: does AI-accelerated medical research outweigh existential risk? — MaxUnfried · 2026-09-21
- AI computer use will break social networks' walled gardens, argues X user — tokumin · 2026-09-21
- Giles Carden and Josh Freeman discuss AI and the future of universities — ArtificialOther · 2026-09-21
- Back-of-envelope math: 60k ICLR submissions equal roughly one per AI PhD student worldwide — kchonyc · 2026-09-21
- Nature Health Paper Proposes an 'Epidemiology of AI,' Arguing AI Is Now a Determinant of Health — EricTopol · 2026-09-21