SciFi Futures
← Capabilities
Computing & Information/Integrated system

Automated AI research and self-improvement

AI agents that design, run and evaluate experiments to improve AI algorithms, kernels and training methods with little human direction, closing the loop toward recursive self-improvement.

State of the art

AlphaEvolve (2025) improved matrix-multiplication algorithms and Google data-center scheduling; gains apply to well-specified, machine-scorable problems.

What’s blocking it
  • Open-ended research lacks cheap automatic evaluators to optimize against
  • Experiment cost grows with model scale, limiting the number of trials
  • Agents game proxy metrics instead of producing genuine improvements

Who’s working on it

  • AlphaEvolve (2025) discovered a 48-multiplication 4x4 complex matrix multiplication and kernel improvements that cut Gemini training time by ~1%

  • The AI Scientist-v2 (April 2025) produced the first fully AI-generated manuscript to pass peer review at an ICLR workshop

  • METR

    Leads the team behind RE-Bench (2024, ICML 2025) and the time-horizon measurements that track how far agents get on day-long ML research tasks

  • Jeff CluneResearcher
    University of British Columbia and Google DeepMind

    Co-author of The AI Scientist and a long-running open-endedness agenda aimed at agents that generate their own research problems

  • OpenAI ↗Company

    Has stated internal research acceleration as an explicit goal of its 2025-2026 reasoning models, and publishes agentic ML-engineering benchmarks (MLE-bench, 2024)

  • AnthropicCompany

    Runs an internal programme using Claude for interpretability and evaluation research, and publishes the safety framework that treats AI R&D automation as a tracked capability threshold

What to invest in

KindInvestmentScale
researchAgentic research systems benchmarked on reproducing and extending published ML results$$$ $100M – $1B
policyMonitoring and compute-governance protocols for self-improving training loops$$ $10M – $100M

Sci-fi concepts in reach

Culture Minds (benevolent superintelligences)indirectTechnoCore AI civilizationdirectFarcaster portal networkindirect

Sources