The Model Is Not the Mind: Rethinking Agent Economics via Erlang/BEAM
arthurcolle · x · 2026-10-07
arthurcolle proposes a 'world + mind' agent architecture built on Erlang/BEAM philosophy: the world is any live-state system (trading book, weather feed, fleet); a mind accretes on it as the model proposes code, the world validates against its own data, and accepted modules fan out. Learning is code-loading; memory is a self-readable catalogue with source and docs. The economics invert — today's frameworks pay tokens per action forever, this pays tokens per capability once, so the system gets cheaper as it lives. Crashes are forgetting, supervisor trees act as immune systems, and the mind outlives its proposer: swap the model, the catalogue stays. The point isn't a bigger model but a world talked to long enough that it no longer needs one.
More from AGI Musings
- Thinker argues economy is cognition-extended nature, and AGI will operate on alien coordination strata — danfaggella · 2026-10-07
- The Fractal Frontier: AI may reveal the world is messier than human math prefers — IgorCarron · 2026-10-07
- Beff Jezos on the unreasonable effectiveness of chain of thought RL — beffjezos · 2026-10-07
- OpenAI's New Math Repo Could Bolster Plasma Research, With Vlasov–Maxwell Result 362 in Focus — johnseach · 2026-10-07
- When math is outsourced to OpenAI, what happens to econ research? — RexDouglass · 2026-10-07
- Writing a precise spec IS half the work — AI helps those who already specify well — sujingshen · 2026-10-07