Not included here, but worth saying: modeling ourselves as in an “AI race” really ceases to make any sense immediately before RSI. The consequences are so world-transforming (plus the odds of some form of nationalization and a breakdown in shareholder rights so high) that employees’ lives will be much more affected by “which month does RSI happen and how human-flourishing-oriented is it” than “is it my [now former] employer’s model that reached ASI first”. Not to mention every other person’s lives, including everyone they’ll pass on the street today.
yes, nonsofic groups exist: this statement is one of many new beautiful results proved by Astra, our next major model.
We're releasing 10 such Astra proofs, complete with lean certificates and CoT walkthroughs for each of them. The results are wide-ranging, from von Neumann algebras (disproof of Connes' Rigidity Conjecture) to better bounds for high dimensional sphere packing, for circuit complexity, for monochromatic triangles in multicolored graphs, and more.
More thoughts here: https://t.co/8SjXONeh38
Smalltalk and HyperCard: often mentioned together, but surprisingly different design philosophies IMO!
The Smalltalk school: everything is the same thing. Minimal, elegant. If you want to mod the graphics system be my guest. Powerful but a bit scary and hard to get started in my own experience.
The HyperCard school: it’s just a friendly drawing app, don’t worry :) oh and you can add links. Oh and you can add buttons. Oh and you can customize the button scripts. Oh whoa you made a game!
Both have benefits, neither is more “right”. Arguably some of their virtues may even be orthogonal.
But I think the smartest programmers tend to underestimate the value of the HyperCard approach.
I was always slow on Forward v.s. Reverse KL. Then I realized a mnemonic trick.
FOrward -> FOMO
Fear of missing out on some probability mass:
min_q KL(p || q) encourages mode coverage of p.
Reverse -> Retreat
Retreat to a high-density mode:
min_q KL(q || p) is mode seeking.
A workflow I'm enjoying: "Walk-driven development"
> go on a nice walk outside 🚶
> record a long audio note: ideas, goals, things to build 🎙️
> agent auto-creates docs/tasks, and kicks off cloud coding agents for me 🤖
Introducing Low-Voltage Inference (LVI) for high throughput workloads.
Today, AI chips can't scale FLOPs without thermal throttling.
As FLOPs utilization increases, AI chips draw more power and downregulate clock speed. This often results in sustained inference throughput under half of peak FLOPs.
Chips in other industries solve the power problem by running at lower voltages. Bitcoin miners run at under 3x the voltage of AI chips!
We’ve designed a new architecture to run our chip’s math blocks at under half the voltage of most AI chips. This enables multiple times the FLOPs density of AI chips today.
We can run trillion parameter sparse MoEs at 80%+ peak FLOPs without thermal throttling.
Running LVI requires co-designing the entire cluster from the transistor to the token: new splittable math arrays, circuit techniques, novel tiling and scheduling algorithms, power delivery networks, VRM architectures, advanced packaging, cold plate designs, and more.
We achieved gold medal-level performance 🥇on the 2025 International Mathematical Olympiad with a general-purpose reasoning LLM!
Our model solved world-class math problems—at the level of top human contestants. A major milestone for AI and mathematics.
OpenAI achieved gold medal on 2025 International Math Olympiad (solving 5 of 6 problems)!
Thinks for hours and writes proofs in natural language.
We've come a long way from LLMs solving 50% of MATH dataset in 2022
Congrats @alexwei_ on spearheading a major milestone!
1/N I’m excited to share that our latest @OpenAI experimental reasoning LLM has achieved a longstanding grand challenge in AI: gold medal-level performance on the world’s most prestigious math competition—the International Math Olympiad (IMO).
This is on the scale of the Apollo Program and Manhattan Project when measured as a fraction of GDP. This kind of investment only happens when the science is carefully vetted and people believe it will succeed and be completely transformative. I agree it’s the right time.
@mollyfmielke There's evidence for it: "In all cases, with exception of S9, they report having owned 1-of-3 toys widely sold by Fisher-Price between 1972 and 1989"
Anecdotally, friend traces some # colors to license plate on family car.
https://t.co/OiyyXKhdkJ
study: https://t.co/aPjthgDW8R