Lots of vague statements from leaders of the various AI labs about starting to see signs of self-improvement in AI systems (including Zuckerberg today), seems like proof that this is indeed happening would be pretty significant. (thanks o3 for providing details & saving me time)
AGI
-
Understanding the Inevitable Problem with AI and Technology
By
–
Everyone sees the problem eventually. They often slam into it at great speed, in fact…
-
Real-world testing essential for genuine AGI validation
By
–
The world has much higher entropy than any benchmark, or set of benchmarks. Therefore no amount of benchmark acing can guarantee your AI has a good world model. The only valid test of AGI is how well your AI does in the real world. No body, no AGI.
-
OpenAI’s General Approach to IMO Gold Medal Achievement
By
–
The most interesting part about OpenAI getting a gold medal in the IMO competition is not really the result itself but rather how it was achieved. Two things were interesting:
— Peter Gostev (@petergostev) 30 juillet 2025
1) OpenAI are deliberately prioritising methods that are general rather than specific;
2) They have… pic.twitter.com/yPsGNyvTUIThe most interesting part about OpenAI getting a gold medal in the IMO competition is not really the result itself but rather how it was achieved. Two things were interesting: 1) OpenAI are deliberately prioritising methods that are general rather than specific; 2) They have
-
Context Rot Term Should Be Used More Frequently
By
–
Context Rot is an excellent term and should be used more often
-
Improving AI Model Training with Enhanced Scaling Approaches
By
–
Good catch! This likely would be fixed with more robust training (more steps + rollouts, and ideally a larger than 1.5B model) 🙂 This was mostly just to show how it works. Feel free to try the same task with a larger model, more steps etc.
-
Dario Amodei on Exponential AI Growth and Scaling Laws
By
–
New interview with Dario Amodei, here is the tl;dr – Exponential growth continues: Amodei firmly believes that the exponential increase in AI capabilities will continue.
– No end in sight: He currently sees no end or slowdown to this trend. The “scaling laws” (more computing -
Innovation Framework: Weekly Capabilities and Emerging Possibilities
By
–
When it comes to what I talk about in videos, it all comes down to these questions… "What can I do this week that I couldn't do last week?"
"What will I soon be able to do that we haven't been able to do before?"
"Will sharing this info genuinely be helpful to a lot of -
CEO Claims Superintelligence, Stock Soars 7 Percent
By
–
This is what happens when, as CEO, you say you are on the verge of achieving superintelligence and discover the first simple processes of self-improvement in AI. Meta stonks up by 7%