VCs rediscovering the trichotomy property of real numbers:

We are starting to see some nuanced discussions of what it means to work with advanced AI In this case, GPT-5 Pro was able to do novel math, but only when guided by a math professor (though the paper also noted the speed of advance since GPT-4). The reflection is worth reading.

Very cool, but I think it’s worth clarifying: (1) it seems that the work here relies on 18 months of work by human mathematicians—it’s not independent, and (2) if I understand correctly what was done here, I don’t think it’s accurate to describe it as “autonomous.”

GPT-5 Pro found a counterexample to the NICD-with-erasures majority optimality (Simons list, p.25). simons.berkeley.edu/sites/default/… At p=0.4, n=5, f(x) = sign(x_1-3x_2+x_3-x_4+3x_5) gives E|f(x)|=0.43024 vs best majority 0.42904.

JP Morgan announced today that they will make direct equity investments of up to $10 billion into strategic industries that support the United States’ economic security. This includes critical minerals. jpmorganchase.com/newsroom/press…

NEWS: Taylor Swift to enter into a multibillion dollar deal with OpenAI to deploy 10 gigawatts of AI data centers

In this new essay, I lay out the framework for what truly spatially intelligent world models must achieve—three essential capabilities for building AI that creates with a storyteller's imagination, navigates with a first responder's fluency, and reasons about space with

Everyone is a math person

The interface of the future is human. We’ve raised a $40M Series B from CRV, Scale, Sequoia, and YC to teach machines the art of being human, so that using a computer feels like talking to a friend or a coworker. And today, I’m excited for y’all to meet the PALs: a new

Modelbusters are reshaping tech. This week, Gamma (70+ million users) raised their Series B, Cursor raised their Series D, World Labs released their spatial intelligence platform Marble, and ElevenLabs crushed the first ElevenLabs Summit. The companies that successfully

Human beings are called to be co-workers in the work of creation, not merely passive consumers of content generated by artificial technology. Our dignity lies in our ability to reflect, choose freely, love unconditionally, and enter into authentic relationships with others.

For the first time in history, over 1 in 3 startups were started solo this year. Introducing The State of Solo Founding: exclusive data from Carta + commentary from top founders and investors. solofounders.com/report Solo founders seem odd today. Soon they'll be the default.

In the late 1830s, Karl Weierstrass dropped out of university. He is said to have spent his school years drinking and fencing. Decades later, he published a function that threatened everything mathematicians thought they understood about calculus. quantamagazine.org/the-jagged-mon…

Terry Tao told me something fascinating: neural networks aren’t grasping ideas the way humans do—they’re excavating structure. They sift through data and surface relationships we didn’t even know to look for. In knot theory, for example, a fairly standard neural network exposed

My unfair advantage is that I love this shit

I left MIT when the cost of ideation became too financially high. I left Anthropic when the cost of loss aversion became too emotionally high. In 2026, every researcher should be asking themselves: where is the frontier today and what is the cost of not being close to it?

Some thoughts on AI and mathematics, inspired by "First Proof."

How are mathematicians facing the wave of rapidly advancing AI-for-math capabilities? Jeremy Avigad (CMU prof and co-author on the original 2015 system description paper for Lean) just posted a paper with his thoughts in the wake of the Math, Inc. announcement on sphere packing.

I moved to SF a month ago. Something shifted immediately. These are sacred grounds. The energy of the people who built the internet, who built the infrastructure that lets one guy with no employees build a platform, a movement, you can feel it. I barely go out. Barely date.

Stanford just tested whether LexisNexis and Thomson Reuters’ AI legal research tools are really “hallucination-free,” as they claim. Spoiler: not even close. Here’s what the study found.