On lab data through September 2026, AI's share of its own research is far from self-sustaining, so a software-only explosion is not near.

groundsWhere's the "intelligence explosion"?, Ramez Naam

Here’s my take: **Given our best current data, the AI self-improvement loop would need to be roughly 5–10× stronger to sustain itself, let alone run away.

qualifiesWhere's the "intelligence explosion"?, Ramez Naam

I might be putting too much weight on a few observations and reaching a comforting conclusion I want to believe.

rebutsWhere's the "intelligence explosion"?, Ramez Naam

Our processes are not ready to handle an intelligence explosion and we appear to be going full-steam ahead toward one.

groundsWhere's the "intelligence explosion"?, Ramez Naam

Here’s Anthropic in the Mythos 5.1 system card: > “we believe that internal usage of recent AI models has been a key factor in **maintaining** the current rate of progress, but we do not yet see clear signs of dramatic acceleration beyond that rate.

groundsWhere's the "intelligence explosion"?, Ramez Naam

At the same time, the graph reports zero cases of AI autonomously completing AI R&D tasks.

backsRecursive Self-Improvement through Multi-Agent Self-Supervision, Hyunin Lee, Jinglue Xu, Jeffrey Seely, Donghyun Lee, Somayeh Sojoudi, Matei Zaharia, Yujin Tang

The evidence for improvement therefore depends on the aggregation unit, and does not establish a consistent gain across research tasks.

backsRecursive Self-Improvement through Multi-Agent Self-Supervision, Hyunin Lee, Jinglue Xu, Jeffrey Seely, Donghyun Lee, Somayeh Sojoudi, Matei Zaharia, Yujin Tang

Mean scores among delivered papers are 3.07, 3.00, and 2.88. These conditional means refer to different sets of completed tasks; they do not measure a quality change on a common completed set. The aggregate score improvement therefore accompanies higher completion.

backsWhat if automating AI R&D triggers an intelligence explosion?, Alan Chan, Christoph Winter, Andrew Barto, Jakub Pachocki, Geoffrey Hinton, Eric Horvitz, Yoshua Bengio, Dawn Song, Jack Clark, Hilary Greaves, Anton Korinek, Samuel Hammond, Thore Graepel, et al.

Productivity gains from AI R&D automation have not yet reached the thresh- old needed to trigger an intelligence explosion, but gains from newer systems are likely approaching that threshold [37].10 The rapid pace of AI R&D automa- tion suggests that this gap will continue to narrow.

backsSource document

Claude is not operating fully autonomously for any measured subset of AI R&D work.

qualifiesWhat if automating AI R&D triggers an intelligence explosion?, Alan Chan, Christoph Winter, Andrew Barto, Jakub Pachocki, Geoffrey Hinton, Eric Horvitz, Yoshua Bengio, Dawn Song, Jack Clark, Hilary Greaves, Anton Korinek, Samuel Hammond, Thore Graepel, et al.

Productivity gains from AI R&D automation have not yet reached the thresh- old needed to trigger an intelligence explosion, but gains from newer systems are likely approaching that threshold [37].10 The rapid pace of AI R&D automa- tion suggests that this gap will continue to narrow.

backsResearch acceleration: The view inside OpenAI, OpenAI

AI research is a complex process with many potential bottlenecks, so the overall pace of progress likely won’t keep pace with these specific metrics.

backsResearch acceleration: The view inside OpenAI, OpenAI

In the last 6 months, over half of successful 4-8 hour tasks involved 1 or more interventions.

qualifiesResearch acceleration: The view inside OpenAI, OpenAI

In terms of a standard 8 hour workday, as of mid-August, in total, the research organization uses 3.1 agent-workdays of effort for every workday of human labor.

qualifiesAn Alien Mind, Jakub Pachocki (OpenAI)

Based on internal results, I have a strong expectation that this speed of progress could be sustained into recursive self-improvement.

backsSource document

We understand that AI is not yet able to make highly significant and fully autonomous contributions to the most important parts of the model training process.