2 hours ago
8
Broadcom (NASDAQ: AVGO) has risen to near the top of the pyramid when it comes stocks that are considered strong artificial intelligence (AI) investments. While it isn't quite to the level of Nvidia, it's likely just a tier or two below it. Broadcom is an enormous company, currently worth around $780 billion.
That means it's a stone's throw away from achieving the illustrious $1 trillion market capitalization, as it only needs to rise just shy of 30% to achieve this impressive feat. But could it do this in 2025? Let's take a look.
Broadcom's growth figure is highly skewed
It isn't easy to summarize Broadcom's work because it does so much. It offers its clients hardware and software solutions in various industries and chip design services. On the hardware side, its networking switches have gotten a lot of attention, as these devices are critical in building out data centers meant to train AI models.
On the software side, Broadcom has products that allow businesses to control their mainframe computers, which is critical in today's increasingly digital business environment. It also has cybersecurity software, but its biggest software product came from an acquisition.
Last year, Broadcom purchased VMware, which allows its users to establish a virtual desktop on the cloud. This has been a successful acquisition, and it's the primary reason why Broadcom grew revenue in its latest quarter.
In its fiscal 2024's third quarter (ended Aug. 4), Broadcom's revenue rose 47% year over year to $13 billion. While that sounds impressive, what's going on under the hood is far less so. If you subtract out the effect of the VMware acquisition (which didn't contribute to the year-earlier results), revenue only rose 4%. That's a substantial change and completely modifies how investors might look at Broadcom.
But that doesn't mean AI isn't helping Broadcom's business. CEO Hock Tan said this during Broadcom's Q3 earnings call:
As you know, our hyperscale customers continue to scale up and scale out their AI clusters. Custom AI accelerators grew three-and-a-half times year on year. In the fabric, Ethernet switching, driven by Tomahawk 5 and Jericho3-AI grew over four times year on year, while our optical lasers and thin dies used in optical interconnects grew threefold.
When you break that down, that's some incredible growth. Connectivity switches grew 400%, while custom AI accelerators, such as Alphabet's Tensor Processing Unit (TPU), which provides better AI performance than an Nvidia GPU, were up 350%!
