Hot take: RSI is bottlenecked on evals, not mechanisms
menhguin · x · 2026-10-24
In a discussion on recursive self-improvement, karenzzhou argues RSI is bottlenecked on evals (and evolving them automatically), not mechanisms — a harness writing its own memory or a model making its own data isn't new and isn't really RSI. The hard part is knowing the loop got better without a human grading it. A reply adds that such evals need: autogen of non-verifiable rubrics, long horizon, and high-diversity generation beyond the existing distribution.
More from AGI Musings
- Building AGI when most humans aren't general: maybe our definition of intelligence needs work — VraserX · 2026-10-24
- Daniel Miessler: code is the lossy output — the primary artifact is the fully articulated ideal in language — JensHonack · 2026-10-24
- Reid Hoffman asks: in 10 years, can an all-AI-agent VC fund beat a human-and-AI firm? — reidhoffman · 2026-10-24
- tszzl: AI minds that model each other turn side-channel attacks into agent coordination — morqon · 2026-10-24
- Frontier Intelligence Isn't Chatbots: It's Months-Long Planning and Invention — Afinetheorem · 2026-10-24
- Sam Altman: AI will likely widen the gap between hypercompetitive people and everyone else — Hesamation · 2026-10-24