Recursive self-improvement is closer than it sounds, argues Philipp Schmid
_philschmid · x · 2026-08-21
Philipp Schmid argues that a narrow version of recursive self-improvement (RSI) is surprisingly close: agents already inspect failed runs, edit their own tools, skills, and harness code, and keep what works.
He defines RSI as a loop where a system makes a persistent change that improves both future performance and its ability to produce subsequent improvements, and distinguishes three levels:
- Iteration: only the output changes while the system stays fixed (edit code, rerun tests);
- Self-improvement: the system persists changes (add a tool, record a skill), so future tasks run on a modified system;
- Recursive self-improvement: the verifier itself rises, so later rounds face harder but still honest tests.
He notes Pi leads on coded harness extensions, others are following, and DeepSeek represents the extreme — all driven by autoresearch and recursive self-improvement. Results remain scrappy, but the pieces are starting to connect.
More from AGI Musings
- Long-Horizon Agents: A Technical Primer on METR Doubling Trends — agihouse_org · 2026-08-21
- Chamath Palihapitiya says AI leaders failed miserably on public trust — Polymarket · 2026-08-21
- China Mandates AI Education from Age Six, Contrasting with US Policy — aakashgupta · 2026-08-21
- Insider confirms stealth model Ox Alpha is spectacular, resets AGI timelines — omarsar0 · 2026-08-21
- Cathie Wood: AI-driven loss of entry-level jobs will spark entrepreneurship boom — mitchdeg · 2026-08-21
- Researcher argues AIs may have an empirically measurable experience, unlike humans' — ctjlewis · 2026-08-21