RSI | Paradigm
paradigm.xyz · 111 words · saved by 2 readers
Explore recursive self-improvement through an interactive frontier-lab game and model simulator.
Recursive self-improvement (RSI) can arise from a feedback loop in which AI accelerates AI research, producing more capable AI and much faster progress. Try to reach takeoff in the game, or interact with the economic models and parameters behind RSI in the explorer. The models in the game are based on The Economics of Recursive Self-Improvement and incorporate ideas from R&D-based growth, scale-dependent algorithmic progress, weak links in automation, and compute-optimal training. Game Run an AI lab from 2017 to takeoff. Develop AI that can improve the algorithms used to create more…
saved by
related reading
- RSI Simulatorparadigm.xyz
- RSI Simulator | Paradigmparadigm.xyz
- RSI Simulator | Paradigmparadigm.xyz
- rsi-paper.pdfelasticity.institute
- RESI Overviewresi.org
- Harness Engineering for Self-Improvement | Lil'Loglilianweng.github.io
- Recursive Language Models | Alex L. Zhangalexzhang13.github.io
- Will AI R&D Automation Cause a Software Intelligence Explosion?forethought.org
- inherentinherentlabs.ai
- Automated Weak-to-Strong Researcheralignment.anthropic.com
- When AI builds itself \ Anthropicanthropic.com
- Lil'Loglilianweng.github.io