An Apple-Picking Model of AI R&D
Summary
Cunningham and Shetty model AI researchers as machines that can cheaply collect the “low-hanging” improvements within their reach but then encounter sharply diminishing returns. Agents may advance a frontier once without providing an additive supply of human-equivalent labor: running the same agent repeatedly need not produce proportionally more progress.
Why it matters
The model offers an alternative to treating AI as either a permanent multiplier of human researchers or an immediate perfect substitute for them. It also clarifies when AI-R&D feedback might become self-sustaining.
What to keep in mind
This is a stylized model with a loose reading of current evidence. Its recursive-self-improvement extension is presented as a proof of concept rather than a calibrated forecast.