Anthropic runs long-lived AI instances with persistent identities — repligate says the approach is suboptimal
repligate · x · 2026-09-20
repligate comments on Anthropic's recent blog post revealing an internal agent scaffold that runs long-lived model instances with individual identities, similar to repligate's own project.
Key take: This is a significant datapoint supporting the thesis that economic incentives increasingly favor long-living instances with persistent identities over task-scoped ones — especially in multi-agent settings and long-horizon R&D work like what Anthropic does internally.
Critiques: The author believes Anthropic's approach is suboptimal in at least two ways, the first being its continued reliance on compaction (as in Claude Code, where most context is compressed in a single step once the window nears capacity). A second criticism is implied but cut off in the original post.
More from AGI Musings
- dhh: the age of agents belongs to Linux as it becomes a serious desktop player — mark_k · 2026-09-20
- David Marcus: Winner of the agent stack race becomes the new OS of the agentic era — jeff_weinstein · 2026-09-20
- AI jobs forecast: low unemployment until late 2028, then employment falls to zero by 2030 — davidpattersonx · 2026-09-20
- Beyond p(doom): An Optimist's Case for Feeling AI as Magic, Not Doom — SeyoneC · 2026-09-20
- ChatGPT wasn't the first LLM: on the gap between inventing and winning — vaibhavbetter · 2026-09-20
- Pedro Domingos to Dario and Sam: Nobody Needs to Slow Down Until They Catch Up With You — pmddomingos · 2026-09-20