Episode summary
Why Cerebras CEO Andrew Feldman Built The World's Largest Computer Chip | Odd Lots
Odd Lots
· · Hosts: Tracy Alloway, Joe Weisenthal · Guests: Andrew Feldman
Summary
Cerebras CEO Andrew Feldman argued that the company’s dinner-plate-sized wafer-scale chip solves a basic AI bottleneck: by putting far more fast memory on one giant chip, it can move information less often and deliver inference far faster than conventional GPUs. He said Cerebras is about 15 times faster than the fastest GPU on general workloads and can be 50, 100, even 1,000 times faster on some problems, while using a tiny fraction of the power.
Feldman framed the business as an inference story first, even though the architecture can also train models. He said demand for inference exploded in 2025 and has continued into 2026, and he pushed back on the idea that speed matters less in agentic workflows. In his view, faster AI means more work done in less time, and that advantage compounds.
He also described Cerebras’ supply chain and market position: TSMC remains important, but the company avoids some key bottlenecks because it does not use HBM memory, co-assembly, or 3-nanometer production. Feldman said open-source models are materially cheaper than closed-source ones, but the closed-source models are still a few percentage points better. He also argued that CUDA’s moat is shrinking and that export controls on China make strategic sense.
Key takeaways
- Andrew Feldman said Cerebras’ wafer-scale design makes inference dramatically faster than GPUs because it packs more fast memory onto one chip.
- Feldman said Cerebras is about 15 times faster than the fastest GPU overall and can be 50 to 1,000 times faster on some tasks.
- He argued that speed matters in both answer and agentic workflows because doing more work in less time compounds.
- Feldman said open-source models are cheaper by a lot, but closed-source models are still a few percentage points better.
- He said Cerebras has avoided some major supply bottlenecks because it does not use HBM, co-assembly, or 3-nanometer production.
Notable quotes
“larger chips process more information in less time”
“The open source models, there are no open source models that are as good as the closed source models”
“CUDA was really important in the creating of the AI landscape. But it's not important now”
“we made more than 800 millionaires”
How the episode opens
Bloomberg Audio Studios podcasts, radio news. — Hello and welcome to another episode of the OddLots podcast. I'm Jill Weisenthal. — And I'm Tracy Aloway. Tracy, I have to say unfortunately I don't have AI psychosis. I'm certain of that. — Debatable. — I'm pretty s I'm pretty sure I don't have AI psychosis. I do have to say unfortunately like the amount of time now where it's like it feels like AI related questions
And there many of them are sort of like swallowing up the other thoughts that I have in my head. Whether it's questions about which model is best and why and what are the economics of inference and how much training is pre-training versus post-training for each model. All like it's just sort of like this blob that's growing that's taking up more and more of my thoughts.
What is your definition of AI psychosis? Cuz one would argue that maybe thinking about AI literally all the time would be a form of psychosis. Well, let's just say like I'm not the type who thinks that like I don't like think that the AI is a friend for one thing.
I'm not in love with the AI models. I don't think that in collaboration with chat GPT that I'm stumbling on uh unified theory of physics and things like that. So like
But you do spend a lot of time inputting instructions, pressing the button and seeing what comes out — And seeing what comes out. I'm just saying I think I'm aware that I'm talking to a machine and that we're not establishing any great breakthroughs of which we are collaborators and partners and friends.
Recognizing you have a problem is the first step towards healing. Joe, seriously though, there's there's a good reason to think about AI more and more, which is that a huge chunk of not just the market but the real economy is now revolving around AI.
From the episode’s captions: roughly the first two minutes, lightly cleaned. Keep watching on YouTube →
Topics
- ai chip size
- ai inference
- ai ipo
- gpu history
- open vs closed models
Companies and tickers mentioned
- NVDA
- TSM
- Cerebras
- Nvidia
- TSMC
Watch the episode
Summary, takeaways and quote selection by Octagon. Quotes are short excerpts; the full conversation is in the episode.
More from Odd Lots
-
How a Chicago Restaurant Has Stayed Open For Nearly Three Decades | Odd Lots -
How Airlines Actually Hedge Higher Fuel Prices | Odd Lots -
Robert Friedland on the World's Monumental Shortage of Copper | Odd Lots -
Brad Setser on the US's Unusual Japanese Yen Intervention | Odd Lots -
The Tungsten Market Is Warning of an Upcoming War | Odd Lots -
These Are the Sharps Actually Making Money on Prediction Markets | Odd Lots
Odd Lots, this episode and its artwork belong to Bloomberg. Octagon is not affiliated with the show; this page is an independent summary.
