These 2 AI Giants Could Soar in 2026 (Hint: It’s Not Nvidia)


Nvidia has been a dominant force in the AI computing market since the AI buildout began in 2023. However, viable alternatives are starting to pop up that could challenge Nvidia’s dominance. If Nvidia’s empire starts to crack, two stocks could soar in its place.

I think both are excellent buys for 2026 and are primed to soar, even if Nvidia can maintain its superiority.

Image source: Getty Images.

Nvidia designs graphics processing units (GPUs), but it doesn’t manufacture them. Instead, it farms out that work to a handful of other companies. In the chip manufacturing world, that work primarily goes to Taiwan Semiconductor (NYSE: TSM). Taiwan Semiconductor is the world’s largest chip foundry by revenue, and it has achieved this position through continuous technological innovation and strong production yields.

Many of Nvidia’s competitors also source their chips from Taiwan Semiconductor, so if Nvidia’s dominance slips, Taiwan Semiconductor will be OK. Additionally, if Nvidia maintains its incredible grip on the artificial intelligence computing market, TSMC will continue to do great. For Taiwan Semiconductor to succeed as an investment, AI hyperscalers need to continue spending at their current rate. If they do, Taiwan Semiconductor will be well-positioned to soar in 2026, as it trades at a discount to the fabless chip companies.

The three primary computing unit providers in the AI arms race are Nvidia, Advanced Micro Devices, and Broadcom (NASDAQ: AVGO). Nvidia is the largest of all of them, and happens to be the cheapest. But none of these three is cheaper than Taiwan Semiconductor.

NVDA PE Ratio (Forward) Chart
NVDA PE Ratio (Forward) data by YCharts

Taiwan Semiconductor trades for 29 times forward earnings. That’s a hefty discount to its peers, despite benefiting from the same tailwinds pushing these other stocks higher. I think this low starting valuation will play in its favor during 2026 and will be a key reason why it outperforms Nvidia in 2026.

Broadcom is taking a different approach to the AI computing problem than its peers. Instead of creating a broad-purpose GPU like AMD or Nvidia, it’s partnering directly with AI hyperscalers to build custom AI accelerators. These computing units aren’t designed to handle a wide variety of workloads like a GPU is. Instead, it’s designed specifically around the workload that it will see. By doing this, Broadcom can maximize performance while decreasing costs, albeit at the expense of flexibility.



Source link