Skip to main content

What Leaders Say: Math in the AI Age

The people building the AI future agree: mathematical thinking is the skill that endures when AI can write your code, draft your emails, and generate your reports.

The academic research backs this up too — see the studies.

For the young, 20-year-old Jensen, that's graduated now, he probably would have chosen more of the physical sciences than the software sciences.

Jensen Huang, CEO of NVIDIA

At the 2024 World Government Summit, Huang argued that programming is no longer a vital skill — AI can handle that. What matters is understanding the math and physics underneath: computational thinking, the laws of physics, mathematical modeling. NVIDIA's own engineers, he notes, need to understand the mathematical foundations of parallel computing more than any specific programming language.

Physics (with math).

Elon Musk, CEO of Tesla & SpaceXreplying to Telegram CEO Pavel Durov's viral "pick math" post

Musk has long argued that solving complex problems requires reasoning from first principles — a mindset rooted in physics and math. While AI can write code for you, it can't replace your ability to think mathematically about problems. He invested in this belief directly: his experimental school at SpaceX led to Synthesis Tutor, an AI-powered math program for kids.

If you learn all the new AI stuff, you're drawing more on your math skills than your coding skills.

Bill Gates, Co-founder of Microsoft

The Gates Foundation made math education its biggest K–12 priority, backed by $1.1 billion in new investment. Gates argues that the way new AI models work is extremely mathematical, and understanding AI means understanding math. He advocates for math teaching focused on real-world problem-solving — budgets, estimation, population growth — not rote calculation.

You can't beat AI on raw horsepower. That's over.

Sam Altman, CEO of OpenAIspeaking to students at the University of Tokyo

Altman compares AI to calculators: "We adapted to calculators and changed what we tested in math class." His point isn't that math doesn't matter — it's that rote computation is automated, so what remains is higher-order mathematical thinking. He emphasizes critical thinking, financial numeracy, and manual math skills as foundational cognitive skills that persist even when tools exist.

Coding is just the language we talk to computers. The skill is: how do I innovate? How do I build something interesting for my end users?

Matt Garman, CEO of AWS

Garman predicts that within two to three years, "authoring Java code" may not be a standalone job. The future developer's role is deconstructing problems, deciding what to build, and coordinating AI agents. He advises focusing on critical reasoning, problem-solving, and creativity — not syntax.

I was interested almost entirely in math and physics.

Dario Amodei, CEO of Anthropicon what drew him to AI research

Amodei's path to founding Anthropic ran through physics and computational neuroscience, not software engineering. He emphasizes that AI's core capabilities — reasoning, inference, pattern recognition — are built on mathematical foundations. For the AI age, he highlights critical thinking and hybrid technical literacy as the skills that endure.

If you master 80 to 90 key models, you can improve your thinking and decision-making abilities tremendously.

Charlie Munger, Vice Chairman of Berkshire Hathaway

Munger saw financial numeracy and mathematical fluency as non-negotiable for rational decision-making. His "latticework of mental models" draws heavily on math, statistics, and probability. Warren Buffett praised Munger's ability to rapidly evaluate complex quantitative problems — the kind of gut-feel numeracy that no spreadsheet replaces.

Computer science is valuable less for programming expertise and more for the process of thinking.

Vinod Khosla, Founder of Khosla Ventures

Khosla advises young people to think from first principles and jump into diverse fields like physics, biology, and finance. He argues that as AI automates specialized tasks, the most valuable skills become generalist abilities: critical thinking, quantitative reasoning, and the ability to learn new domains rapidly.

We had faith in Moore's Law, and we had faith in math.

Fei-Fei Li, Professor at Stanford, Co-founder of AI4ALLon building ImageNet, the dataset that launched modern AI

Li studied physics at Princeton before pioneering computer vision. Her work on ImageNet — the dataset that sparked the deep learning revolution — was grounded in mathematical confidence. She emphasizes that AI research requires deep mathematical foundations, not just programming skills.

Computational thinking is a problem-solving mindset that allows students to flexibly adapt to new challenges.

Sundar Pichai, CEO of Google

Pichai, who excelled in science and mathematics before studying at IIT, committed Google to the White House AI Education Taskforce. Google's $1 billion education investment prioritizes not just AI literacy but the mathematical and analytical thinking that underlies it. The emphasis is on fostering problem-solving skills, not teaching specific tools.

The shared message:AI can write code, generate text, and automate tasks. But it can't replace your ability to judge whether the numbers it produces are right. Mathematical thinking, domain intuition, and quantitative reasoning are the durable skills that every leader on this list is betting on.

Build the skill they're talking about

MathMathMath builds numeracy through practice — no multiple choice, no hints. Type the answer. Know it for real.