- Broadcom revenues are surging due to demand from hyperscalers for its AI chips
- It said for 2027 it has secured the supply to again double AI revenue to approximately $115 billion
- In 2028 it has line of sight for its AI semiconductor revenue growth to again double to $230 billion
Broadcom reported some amazing numbers for its third quarter 2026, including revenue of $29.6 billion, up 86% from the prior year period. And it gave fourth quarter 2026 revenue guidance of about $34.8 billion, an increase of 93% from the prior year period.
"Demand for our custom AI accelerators and networking continues to be very strong,” said Hock Tan, president and CEO of Broadcom. Its Q3 AI semiconductor revenue of $16.7 billion grew 221% year-over-year. For Q4, Broadcom expects its AI semiconductor revenue to accelerate to $21.7 billion, up 236% year-over-year.
Tan talked about its AI business possibly doubling next year, driven by demand for its custom accelerated processing units (XPUs).
“We are continuing to see exponential growth in demand from our XPU customers,” he said. “We believe the vast majority of compute demand for AI workloads today originates from this concentrated group who develops state-of-the-art frontier models, and we expect their need for compute infrastructure to inflect even more in 2027 and 2028. And they are all growing with XPUs to achieve superior performance, cost and power.”
Tan said, “Anthropic is on track to become our largest XPU customer in 2027 and sustain that in 2028.” Its other big customers include Google, OpenAI and Meta.
AI semiconductor revenue guidance
The company gave some general guidance for its AI semiconductor revenue in 2027 and even 2028.
It said in 2027 it has secured the supply to again double AI revenue to approximately $115 billion. In 2028 it expects the trajectory of growth to continue. “We have line of sight for fiscal 2028 AI semiconductor revenue growth to again double to $230 billion," said Tan.
Analysts on the earnings call were very interested in Broadcom’s Tomahawk 6 chips. These are high-bandwidth Ethernet switch chips designed primarily for AI and high-performance data-center networking.
Analyst Harlan Sur with J.P. Morgan, said “We've heard that Tomahawk 6 ramp has been the fastest ramp of any of your switching families, and we've also heard that you guys are almost sold out of Tomahawk 6 for next year.”
Charlie Kawwas, president of Broadcom's Semiconductor Solutions, said, “Our Tomahawk 6 has been a phenomenal success. It's available both in 100 gig and 200 gig, SerDes. So we actually have two versions of Tomahawk 6. Both are very successful, and both are deployed in pretty much all of the AI hyperscalers that are building XPUs with us. And even those that are not using our XPUs are using the Tomahawk 6, both 100 gig and 200 gig.”
Kawwas added that its customers who build XPUs with it, deploy either Tomahawk 6 or the company’s newest switch chip Tomahawk Ultra.
“We're seeing scale-up solutions adopt both Tomahawk 6 and Tomahawk Ultra,” said Kawwas. “And the beauty and the reason why they're doing this is it's Ethernet-based, which means it's open, anybody can connect to it, especially with the standards that we have collaborated with the entire industry on. So at this point in time, we're seeing it pretty much deployed in both XPU clusters and in some GPU clusters.”
Read more about Broadcom and AI on Fierce Network
Broadcom sees hyperscalers sidelining telcos for some scale across projects
Nvidia’s latest AI trick: turning storage into memory relief
Facts Only
* Broadcom reported $29.6 billion in revenue for the third quarter of 2026, an 86% increase from the prior year.
* Fourth quarter 2026 revenue guidance is approximately $34.8 billion, an increase of 93% from the prior year.
* Q3 AI semiconductor revenue was $16.7 billion and grew 221% year-over-year.
* Broadcom has secured supply to double AI revenue to approximately $115 billion by 2027.
* The company has line of sight for fiscal 2028 AI semiconductor revenue growth to double to $230 billion.
* Q3 AI semiconductor revenue was $16.7 billion, growing 221% year-over-year.
* Broadcom expects AI semiconductor revenue to accelerate to $21.7 billion in Q4 of 2026, up 236% year-over-year.
* The Tomahawk 6 chips are high-bandwidth Ethernet switch chips for AI and data-center networking.
* Customers building XPUs use either Tomahawk 6 or Tomahawk Ultra switch chips.
* The adoption of Tomahawk 6 and Tomahawk Ultra is driven by the Ethernet-based, open nature of the technology.
Executive Summary
Broadcom is experiencing surging revenue driven by demand for its AI chips from hyperscalers. The company has secured supply to potentially double AI revenue to approximately $115 billion by 2027, with a line of sight for AI semiconductor revenue to reach $230 billion in 2028. In the third quarter of 2026, Broadcom reported $29.6 billion in revenue, an 86% increase from the prior year. Fourth quarter 2026 revenue guidance is set at about $34.8 billion, representing a 93% increase from the prior year.
The demand for Broadcom's custom AI accelerators and networking components remains strong. In Q3 2026, AI semiconductor revenue reached $16.7 billion, growing 221% year-over-year. The CEO noted exponential growth in demand from customers developing frontier AI models, specifically mentioning Anthropic, Google, OpenAI, and Meta as major consumers of custom accelerated processing units (XPUs).
Broadcom's Tomahawk 6 chips are noted for their success in high-bandwidth Ethernet switching for AI and data-center networking. Customers utilizing these chips deploy them across XPU clusters and some GPU clusters, leveraging the open, Ethernet-based nature of the technology.
Full Take
The narrative centers on the immense leverage AI compute demand provides to Broadcom, positioning the company not just as a component supplier but as an essential infrastructure provider for frontier model development. The projections—doubling AI revenue to $115 billion by 2027 and $230 billion in 2028—suggest that the growth is fundamentally tied to the continued concentration of compute demand among leading AI labs, specifically those developing state-of-the-art models. This creates a feedback loop where the performance demands of these frontier models necessitate the specialized interconnects provided by Broadcom's custom chips and networking solutions.
A key tension lies in the deployment strategy: the adoption of Ethernet-based switching solutions like Tomahawk 6 across both XPU and GPU clusters suggests an appeal to systemic interoperability and openness. This pattern implies that infrastructure choices are being dictated less by singular vendor preference and more by the need for scalable, multi-cluster environments, which leverages the 'open' nature of the underlying standards as a competitive advantage. The implication is that the bottleneck shifts from raw chip manufacturing capacity to the ability of hyperscalers to integrate and scale these high-speed fabrics efficiently.
The structure suggests a transition from specific product success (Tomahawk 6 adoption) to broad financial projections, suggesting that incremental hardware advancements are being framed within an exponential growth trajectory driven by concentrated compute needs. The lack of focus on capacity constraints or potential market saturation in the long-term projections warrants scrutiny regarding whether the projected doubling reflects actual demand expansion or merely optimistic extrapolation of current consumption patterns.
Bridge Questions: What mechanisms exist to ensure that this anticipated double-AI revenue is realized across all specified customer segments, and what are the implied cost structures for hyperscalers adopting these massive infrastructure expansions? How does the focus on open Ethernet standards mitigate potential vendor lock-in concerns when specific custom chips like XPUs are central to the demand curve? What factors might introduce volatility into the projected growth trajectory between 2027 and 2028?
Sentinel — Human
The text reads like a summary or excerpt from a formal earnings call transcript, exhibiting the structured data presentation typical of financial journalism rather than pure generative prose.