Researcher's tinfoil theory: bad ML papers may slow AI self-improvement below math/code pace

RylanSchaeffer · x · 2026-09-28

Rylan Schaeffer offers a new "tinfoil hat" theory for why recursive self-improvement (RSI) might arrive slower than math and code benchmarks suggest:

The implication: an AI automating AI research would rely on a corpus far noisier than math or code, dragging down the pace of self-improvement.

Related event: Researcher floats hypothesis: noisy AI literature may slow recursive self-improvement(2 posts)→

Original post →

More from AGI Musings

AGI Musings channel →