Cerebras shares plunged after the company gave an annual sales forecast that disappointed investors who were expecting the company to carve out a bigger slice of the AI data center market. Cerebras CEO Andrew Feldman joins Ed Ludlow to discuss the results and what's next for the company
See omnystudio.com/listener (https://omnystudio.com/listener) for privacy information.
Bloomberg Talks curates top interviews from around Bloomberg News. Hear conversations with the biggest names in finance, politics and entertainment. On Bloomberg Talks, we round up interviews with Fortune 500 CEOs, governme...
What Andrew Feldman said
Written from the verified transcript and checked against it. Every figure links to the moment it was said.
Cerebras CEO Andrew Feldman discussed the company's first quarterly earnings since its IPO, addressing the 17.5% stock drop despite beating sales estimates. He highlighted record revenues of $191 million, up 92% year-over-year, and cloud business growth of 167%. Feldman explained that Q2 and Q3 margins would be impacted by 10-15 points due to renting back gear to customers to meet demand. He emphasized data center capacity as the main constraint, citing a 120 MW partnership with Bell Canada for 2027. He touted Cerebras's architecture, which avoids HBM, CoWoS, and 3nm node constraints, enabling fast deployment, exemplified by signing OpenAI on December 24th and full production by February 1st. He argued the market will be heterogeneous, not GPU-dominated. With over $9 billion on the balance sheet, he remains open to capital markets. Success is measured by beating aggressive plans.
Key takeaways
Record revenue of $191 million, up 92% year-over-year, and cloud business up 167%.
Q2 and Q3 gross margins will be impacted by 10-15 points due to renting back gear to customers.
Full-year gross margin guidance is 10 points above consensus.
Signed OpenAI contract on December 24th and reached full production by February 1st.
Over $9 billion on the balance sheet, scanning capital markets for growth opportunities.
0:10And for me, this is about ramping output for two big customers. Is that true?
2:14Have the proceeds from the IPO actually allowed you to move more quickly in ramping up capacity?
3:10How difficult right now is it to get moving in America or other markets to get planning approval, get the concrete, get the labor, get the thing built?
3:57Would you just explain the basics of the technology, but how insulated are you from the memory bottleneck that others are experiencing?
5:16... you say hand on heart, not just in winning business, but have you actually been able to go to a customer and say, 'We can get this compute online faster than others for those reasons you just outlined,' and then actually gone and done it?
5:48How do you interpret the frontier labs going to custom ASICs alongside their other compute options?
6:53Andrew, if the world is simple and it's money that greases the wheels to get your industry going, can we see you come back to capital markets in some form just so you can move quicker?
7:23What is the milestone that you'd pitch to the market to keep closest attention to?
Cerebras reported quarterly earnings for the first time since going public last month. Its sales outlook beats Wall Street estimates, but still disappointed investors hoping to see the company carve out a bigger slice of the AI data center market. Right now shares are down 17.5%, its biggest drop in its quite short history as a public company. What's behind that? CEO Andrew Feldman is with us. Welcome back to Bloomberg Tech, Andrew. You know, there was a time where we would talk about the merits of top-to-tail server ownership, how owning all of the content. Now we're going to talk about margin contraction and we're going to talk about the stock being down 17%. That to me is kind of the mismatch. The outlook on the sales side beat Wall Street estimates. I think a lot of people are trying to understand the sequential margin decline. And for me, this is about ramping output for two big customers. Is that true?
Yeah, I think what we did is we put forward a plan in the start of '26. We shared it with investors as we went public. And we're ahead of plan. You know, we delivered record revenues of 191 million, 92% year-over-year. And for our cloud business, you know, it was up 167% year-over-year. We beat margin consensus substantially. And then we guided for full year that gross margins would be 10% better than plan. We also shared that in Q2 and Q3 we would go back to some of our customers and we would rent back gear that we'd sold them to try and keep up with demand. And that would have a margin impact on the order of 10 or 15 points. We did that to keep our customers close, to be sure we could keep up with their extraordinary demand for our product for fast inference. And so that was the story. On every metric we put out, we're ahead of plan.
Yeah, I think capacity is the largest constraint right now for everyone. Data centers are. And we've significantly increased our ability and our pipeline for data centers, which is now very large. You know, we announced a data center partnership with Bell Canada for 120 MW. That will be delivered in 2027. We are pursuing data centers across the US and Canada, in Europe, in the Middle East. All the vast resources that we have now at our disposal give us tremendous advantage in the pursuit of this limiting factor, data centers.
What you're talking about matter-of-factly is buildings, not necessarily the compute, right? It's not what you guys are offering. How difficult right now is it to get moving in America or other markets to get planning approval, get the concrete, get the labor, get the thing built?
Well, that's the irony of this market, that the AI market is moving at blistering speed, and we are being constrained by data centers, which move with the speed of real estate. And so that is a problem that is being confronted by everybody in the category, by the hyperscalers, by the neo clouds, by the new generation clouds, everybody is confronting the similar problem.
Andrew, Cerebras does not rely on traditional off-chip HBM. Would you just explain the basics of the technology, but how insulated are you from the memory bottleneck that others are experiencing?
Yeah, that's a really good point. Because of our innovative architecture, because of our wafer-scale approach, we don't use HBM. HBM is a type of DRAM, and it's made by three companies, one of whom is reporting shortly, right? That's Micron, Hynix, and Samsung. There's a global shortage, it's extremely expensive, lead times are long, and we don't use it. So, we have a tremendous advantage there. The other constraints in the supply chain for many are CoWoS, which is a process inside of TSMC. Again, we don't use it. And the third is capacity at the 3-nanometer node. That's space in TSMC's factory that makes 3-nanometer chips. Again, we don't use it. We're at the 5-nanometer node. So, our architecture has allowed us to deliver the fastest inference in the world by an order of magnitude while avoiding the main supply chain constraints faced by others in the field.
9 more exchanges in this transcript
Sign in free to read the rest of this interview. No card required.
Feldman, A. (2026, June 24). Cerebras CEO Andrew Feldman Talks Earnings | Bloomberg Talks [Interview transcript]. Bloomberg Podcasts. CEOInterviews.AI. https://ceointerviews.ai/interview/1353265/
MLA
Andrew Feldman. "Cerebras CEO Andrew Feldman Talks Earnings | Bloomberg Talks." Bloomberg Podcasts, 24 Jun. 2026. Transcript, CEOInterviews.AI, https://ceointerviews.ai/interview/1353265/.
BibTeX
@misc{feldman2026_1353265,
author = {Andrew Feldman},
title = {Cerebras CEO Andrew Feldman Talks Earnings | Bloomberg Talks},
howpublished = {Interview transcript, Bloomberg Podcasts. CEOInterviews.AI},
year = {2026},
month = {jun},
url = {https://ceointerviews.ai/interview/1353265/},
note = {Speaker-attributed transcript with timestamps}
}
