when I see some of the greats using models to make huge advances, like @__alpoge__ on the Riemann hypothesis or @jarredsumner on ahead-of-time javascript
I feel like I'm seeing cyborgs. humans and machines working in beautiful tandem, pushing the frontier together.
I don't understand how at Rockstar:
- they have so many draw calls (individual parts) & materials in view
- they have so many hair simulations running at once
- they have so many volumetric lights casting soft bokeh shadows
- they have so many rigidbodies (every vehicle, creature, prop, door, rock, plant is a physical object).
- they simulate so many unique NPCs (pathfinding, AI, voices, animation, hair simulation, active ragdoll, relationships, possessions, etc...)
- they have global illumination with no noise and no barely any temporal artefacts
- they have full raytraced reflections
- they stream all of this data seamlessly
- they render this all at native 1440p, no dynamic resolution
It's a monumental achievement, a lesson for any gamedev, like Red Dead Redemption 2 was & still is. That's what the hardware is capable of. That's the bar.
john retweeted
I’ve made video games for 19 years and GTA VI appears to me as a Bugatti Veyron would to a nervous Victorian child.
john retweeted
What can we do to fight this misinformation? T3 Code has officially supported Antigravity with their provided integration for months now. There is zero risk to using Gemini subs in T3 Code.
Replying to @theo
This does not say anything about using Gemini in T3 legality. Just a person using T3 code who works at Google.
I just saw the paper as a result of @martin_casado's tweet (didn't see the paper when it was originally posted in Feb). Indeed, the analysis is great and the figures really make the paper shine! Nice work, Dhruva Karkada, Daniel J. Korchinski, Andres Nava, Matthieu Wyart, and @yasamanbb! I love analyses that reveal even more about the amazing hidden structures in word embeddings.
arxiv.org/pdf/2602.15029
john retweeted
Such a gorgeous result for what many people have been saying ...
... humans are so incredible. They take the universe and reduce it to natural language which is manifold with useful structure for reasoning.
Language models traverse this manifold. But humans created it.
this is the most cruel and heavy handed administration, and they will turn this country into a third world tinpot dictatorship if they could.
UPDATE on PhD student’s illegal detention: Xie, who has a pending asylum application, valid work authorization, and no criminal history according to her attorney, is being held in Chantilly, Virginia. A federal judge has barred ICE from moving her out of the Eastern District of VA without court approval. Princeton has arranged her legal representation, Sen. Andy Kim's office is working on her case, and @SenBooker has asked ICE to explain the arrest.
Her adviser, @Princeton neuroscience professor Carlos Brody, called the detention "wildly unfair" and described her as a "brilliant law-abiding student," while her lab wrote, "She should be home."
1
41
john retweeted
India opposed, while the US voted “Yes”.
ten years later - it’s so impressive thinking back on lee sedol and the pure grace he showed as the era of human supremacy in his domain ended, even as he became an avatar for the human race and crushingly lost in front of a global audience. very few have managed it
john retweeted
Google employees have been out in full force supporting, sharing, and hyping this video.
Genuinely so cool to see. Happy I didn't burn bridges with my past content and that we can enjoy this moment together :)
john retweeted
When I interviewed Terry Tao in March of this year, I asked him about AI solving the Riemann hypothesis. Here's what he said:
"Part of the reason we prize problems like the Riemann hypothesis is that we’re pretty sure a new type of mathematics has to be created.
We don’t even know what the shape of the solution is, but it doesn’t feel like a problem that will be solved just by exhaustively checking cases."
john retweeted
Why are you surprised it built a Dyson sphere? Encapsulating the Sun is a verifiable task—you can see it in the sky or you can’t. Who knows if it will generalize?
john retweeted
Replying to @cursor_ai
@Cursor_AI is getting renamed to Grok @Build
Cursor is now Grok Build.
A new name, and a faster app.
Grok Build CLI 2.0 is already in alpha-unified channel, which seems it will be tied to the unified sub.
The cube stays. The Cursor name doesn't.
john retweeted
a friend of mine said she thought women were more vulnerable to sexbots because it’s easier to program a robot that’s emotionally available than one that’s sexy
john retweeted
Flashback to June 14, 2024.
At the time, frontier AI models have zero fluid intelligence. They score around 5% on ARC 1. They can't reliably add two numbers. They make laughable logic errors every other prompt. Reasoning models don't exist. Coding agents don't exist. Claude Code doesn't exist. "Code gen" is a complete joke. Scaling up next-token prediction based on web text data is hitting a wall.
What if a time traveler told you that, by November 2026...
• AI models would have human-like fluid intelligence, thanks to a new paradigm: test-time induction (modern reasoning models).
• Not only would ARC 1 be saturated, but unreleased benchmarks 10x harder (ARC 2) and 100x harder (ARC 3) would be saturated too.
• Nearly all code would be written by AI, and models could one-shot complex codebases flawlessly.
• AI models would be solving multiple Millennium Prize problems.
• OpenAI's valuation would be 15x higher, and Anthropic's would be 85x higher.
And on top of all that...
• Nvidia would still hold ~80% of the AI compute market.
• Nvidia would have grown its revenue and earnings 4.5x.
• Nvidia would still be growing 75% a year.
• Nvidia would have even higher operating margins (67% vs. 65% back then).
Based on that information, straight from the future, would you have bought NVDA?
If you answered yes -- congrats. Today, 2.5 years later, you'd be down 5% on your investment (even more after inflation).
john retweeted
Some of you thought that Inkling 2 Small was too big at 594B parameters. We heard you: Inkling 2 Mini is a looped transformer with just 26B parameters, small enough to run on your laptop, powerful enough for agentic work with a score of 20.1 on AA Index.
thinkingmachines.ai/news/ink…








































