# Recursive Criticality of AI Self-Improvement

arXiv: https://arxiv.org/abs/2609.00137 · HTML: https://mlsanity.com/paper/2609.00137/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 13 of its 34 references that are in the map.
- First posted on arXiv 31 Aug 2026.

## Builds on (most cited first, 13 of 13)

- [PaperBench: Evaluating AI's Ability to Replicate AI Research](https://mlsanity.com/paper/2504.01848/) (2025)
- [Measuring AI Ability to Complete Long Software Tasks](https://mlsanity.com/paper/2503.14499/) (2025)
- [AI Research Agents for Machine Learning: Search, Exploration, and Generalization in MLE-bench](https://mlsanity.com/paper/2507.02554/) (2025)
- [Thousands of AI Authors on the Future of AI](https://mlsanity.com/paper/2401.02843/) (2024)
- [A Rosetta Stone for AI Benchmarks](https://mlsanity.com/paper/2512.00193/) (2025)
- [Recursive Self-Improvement in AI: From Bounded Self-Refinement to Autonomous Research Loops](https://mlsanity.com/paper/2607.07663/) (2026)
- [Algorithmic progress in language models](https://mlsanity.com/paper/2403.05812/) (2024)
- [Self-Modification of Policy and Utility Function in Rational Agents](https://mlsanity.com/paper/1605.03142/) (2016)
- [Measuring AI R&D Automation](https://mlsanity.com/paper/2603.03992/) (2026)
- [On the Origin of Algorithmic Progress in AI](https://mlsanity.com/paper/2511.21622/) (2025)
- [A Mathematical Framework for AI Singularity: Conditions, Bounds, and Control of Recursive Improvement](https://mlsanity.com/paper/2511.10668/) (2025)
- [From AGI to ASI](https://mlsanity.com/paper/2606.12683/) (2026)
- [Stability and performance analysis of linear positive systems with delays using input-output methods](https://mlsanity.com/paper/1703.00405/) (2017)

## Abstract (arXiv)

AI is increasingly used in the R&D process that produces future AI systems. We study the conditions under which this feedback becomes self-amplifying. Our model describes how the rate of AI capability growth depends on baseline research productivity, recursive feedback, and the increasing difficulty of research progress. We derive a recursive reproduction number, RAI, that determines whether improvements are amplified or damped across development cycles. This quantity compares the strength of feedback with the rate at which further progress becomes more difficult. When RAI>1, the effects of improvements compound across development cycles, placing the system in a self-amplifying regime. When RAI<1, their effects weaken across cycles. The transition depends on the structure of the AI R&D feedback loop and need not occur at any particular level of model capability. A system can therefore enter a self-amplifying regime before acceleration becomes visible, while rapid progress can also occur without self-amplification. Higher baseline research productivity can accelerate progress without changing whether the system is self-amplifying, but the duration of the development cycle becomes a limiting timescale for amplification. Increasing research difficulty can end a period of self-amplification. Extending the model to multiple research actors shows that improvements shared across organizations can make the overall research ecosystem self-amplifying even when no individual actor is. The framework identifies measurable properties of AI R&D systems that can help distinguish recursive amplification from rapid progress driven by other sources, including the strength of recursive feedback, how effectively improvements propagate into successor systems, cycle duration, and the increasing difficulty of further progress.