As Broadcom (AVGO) prepares to report earnings next week, Bernstein Managing Director and Senior Analyst Stacy Rasgon appears on Market Domination to share her bullish view on the company.
Rasgon explains that Broadcom has many different business segments, from semiconductors and software to infrastructure. “And I think they have the second best AI story in the space after Nvidia,” he adds. He details how Broadcom is partnering with tech giants like Google (GOOG, GOOGL) and Meta (META) to build and scale its own chips.
He added:[AI] “Frankly, networking is just as important as computing. And networking is the 800-pound gorilla there,” he said, explaining that Broadcom's current outlook for AI revenue this year is more than $10 billion, about a third of its entire semiconductor business.
For more expert insights and the latest market trends, click here to watch this entire episode of Market Domination.
This post was written by Melanie Leal
Video Transcript
Let's change the topic.
Well, my last question is about Broadcom, which will be reporting next week.
You recently commented that this is good for people who aren't that familiar with Broadcom.
Where does this fit into this whole new world of AI, and why are you excited about this?
Ah, yes.
Broadcom does a lot of different things.
They're in the semiconductor business, they're in the software business, and they do a lot of infrastructure work.
They also have what I would call the second best AI story in the space behind NVIDIA, you know, one of the competitive factors for NVIDIA is actually like large hyperscalars, whereas Google and the metas of the world are actually developing their own chips and in most cases relying on other external partners to make it happen.
Broadcom is one of those partners; Broadcom makes AI chips for Google, and Google calls the company TP US.
They have been working with Google for 8-9 years now.
They're also working on the meta.
They are working with a third client, which they revealed recently.
Uh, they also do AI networking, um, Ethernet for large scale language models and generative AI. And networking is frankly just as important as computing, and they're the 800 gorilla.
And the company's current guidance for AI revenue this year is more than $10 billion of the $30 billion in semiconductor revenue it's aiming for this year.
That means this represents one-third of the company's semiconductor business.
Well, they're looking at what's called AI-related stuff this year.
Again, everyone in the AI industry believes that AI is one of the few end markets where it's doing very well.
Hmm, all of these people have an upward bias on that number and the 10 billion number feels pretty conservative to me, but this is what I would call the second best AI story in the space, as I said, behind NVIDIA.
