Framing automated science as a hypothesis-experiment-refute loop, with agent swarms competing
burny_tech · x · 2026-09-23
The author sketches an approximation of the scientific method suited for automation: define a domain, pose a question, gather existing data/models, formulate the most predictive hypothesis your compute allows, test it with experiments, then refine or discard under falsification—and loop. Many steps are automatable. For more open-ended tasks he suggests evolutionary/novelty-search style approaches, even Feyerabend-style methodological anarchism, and extensions like competing swarms of agents exploring rival hypotheses or paradigms. A thoughtful take on the design space of AI-driven empirical science.
More from AGI Musings
- Mathematician Elliot Glazer argues OpenAI should "slop drop" all its math results rather than hide them — burny_tech · 2026-09-23
- Grady Booch doubts AI's Navier-Stokes claim: insights may come from human experts — Grady_Booch · 2026-09-23
- Grady Booch: Contemporary AI Still Lacks Abductive Reasoning, Just 'Next-Token Prediction' — Grady_Booch · 2026-09-23
- AI solves Navier-Stokes-related problem as machines upend mathematics, New Scientist reports — burny_tech · 2026-09-23
- Mathematician says OpenAI likely proved a significant partial case of the Hodge conjecture — burny_tech · 2026-09-23
- An 11-minute billiards video hailed as the best intuitive explainer of LLM behavioral stability — ctjlewis · 2026-09-23