Broadcom (NASDAQ: AVGO) is one of the most important companies in the artificial intelligence (AI) infrastructure space, designing custom processors and networking components for major hyperscalers and pure-play AI companies.
The company’s growth trajectory has improved remarkably in recent years, and its strong financial performance has been handsomely rewarded on the stock market. Broadcom stock has appreciated 265% over the past three years, outpacing the 222% jump in the PHLX Semiconductor Sector index. This impressive rally has made Broadcom one of the world’s largest companies.
Missed Nvidia in 2009? This Rare Signal Is Flashing Again. In 2009, a “Double Down” signal flashed for a little-known chipmaker called Nvidia. For the first time in years, that same “Total Conviction” signal is flashing for a company 1/100th the size of Nvidia. Continue »
Broadcom currently has a market cap of just over $1.5 trillion, making it the 10th-largest company globally. However, I won’t be surprised to see Broadcom becoming significantly larger by the end of the decade. Let’s look at the reasons why this semiconductor stock could triple from current levels and achieve a $4.5 trillion market cap by 2030.
Broadcom is benefiting from two major catalysts
Broadcom designs custom AI processors tailored to perform specific tasks for Alphabet‘s Google, Meta Platforms, Anthropic, OpenAI, and a couple of other customers. These customers have been ramping up deployment of Broadcom’s custom processors to reduce the total cost of ownership (TCO) and energy consumption in their AI data centers.
Broadcom’s custom AI processors have an advantage over general-purpose computing chips, such as graphics processing units (GPUs), because they are purpose-built for specific tasks. As a result, the power consumption of custom AI chips is reportedly just 10% of what a GPU consumes, and they are reportedly three times faster at performing the task they are designed for.
Not surprisingly, shipments of Broadcom’s custom AI processors, which the company calls XPUs, increased by 3.5x year over year in the third quarter of fiscal 2026. On the other hand, Broadcom also designs networking components, such as Ethernet switches and high-speed optical interconnects. The company is experiencing exponential growth in this segment as well, with its AI-focused networking revenue jumping by 2.5x year over year in fiscal Q3.
It is easy to see why that’s the case. Optical networking is emerging as the next big bottleneck in the AI infrastructure space. That’s because AI has created the need to transfer enormous data sets quickly and seamlessly between data centers and chip clusters with low latency, ensuring training and inference workloads can run without hiccups.
