The Best AI Semiconductor Stock to Buy for 2026, According to Certain Wall Street Analysts (Hint: Not Nvidia or Broadcom)

Nvidia (NVDA 0.46%) and Broadcom (AVGO 1.07%) are cornerstones of the artificial intelligence trade. But while Morgan Stanely analysts led by Joseph Moore have buy ratings on both chip stocks, Micron Technology (MU 2.47%) is their top pick within in the semiconductor industry for 2026.

Readers should be aware that most Wall Street analysts have a different opinion.

• Among 69 analysts, Nvidia has a median target price of $250 per share. That implies 33% upside from its current share price of $187.
• Among 52 analysts, Broadcom has a median target price of $460 per share. That implies 31% upside from its current share price of $350.
• Among 44 analysts, Micron has a median target price of $305 per share. That implies 4% upside from its current share price of $293.

Here’s what investors should know.

1. Nvidia

Nvidia is well known for its best-in-class graphics processing units (GPUs), chips used to speed up compute-intensive data center workloads like artificial intelligence (AI) training and inference. However, the company is truly formidable because of its full-stack strategy that spans adjacent hardware and software development tools.

Nvidia develops rack-scale systems that pair GPUs with CPUs and networking equipment, which lets the company optimize performance across the entire data center, while saving clients the trouble of integrating hardware from multiple suppliers.

In addition, Nvidia’s CUDA software platform provides developers with an extensive suite of code libraries, frameworks, and pretrained models that simplify application development across numerous of use cases, from predictive analytics and content generation to conversational agents and autonomous machines.

Consequently, while custom AI chips designed by rivals like Broadcom may be cheaper, Nvidia systems often have the lowest total cost of ownership (TCO) when every expense is include. “Our TCO is so good that even when the competitors’ chips are free, it’s not cheap enough,” according to Nvidia CEO Jensen Huang.

Nvidia dominates the AI accelerator space today with more than 80% market share, and the company is likely to maintain its dominance for the foreseeable future because it offers customers the lowest TCO. Wall Street expects Nvidia’s earnings to grow at 37% annually in the next three years. That makes the current valuation of 46 times earnings look attractive.

2. Broadcom

Broadcom is a key player in two parts of the AI supply chain: networking chips and custom accelerators. The company has 80% market share in high-speed Ethernet switching and routing chips, a market forecast to grow at 20% to 30% annually in the next few years, per JPMorgan Chase. Its latest Tomahawk 6 (switching) and Jericho 3 (routing) chips offer industry-leading performance.

Broadcom is also the leading supplier of custom AI accelerators, also called application-specific integrated circuits (ASICs), with 70% to 80% market share. The company has five major ASIC customers: Alphabet‘s Google, Meta Platforms, TikTok parent ByteDance, OpenAI, and Anthropic. However, Broadcom has other prospective customers in the pipeline, including Apple, Arm Holdings, and xAI. The AI accelerator market is projected to expand at 29% annually through 2033, according to Grand View Research.

Beyond semiconductors, Broadcom subsidiary VMware is a leader in virtualization software. Consultancy Gartner recently recognized the company as a leader in distributed hybrid infrastructure, a setup that uses virtualization to combine public and private clouds into one manageable system. The data center virtualization market is forecast to grow at 21% annually through 2033, according to Grand View.

Wall Street estimates Broadcom’s adjusted earnings will grow at 36% annually during the next three years. That makes the current valuation of 51 times earnings look attractive.

3. Micron Technology

Micron develops memory and storage solutions for personal computers, mobile devices, data center servers, and automotive systems. The company sells DRAM (dynamic random access memory) products, including high-bandwidth memory (HBM) created by stacking DRAM chips, and NAND products.

DRAM products offer fast, short-term memory for active applications, and NAND products provide slow, long-term memory for file storage. Both are important for artificial intelligence workloads; DRAM (especially HBM) products offer the bandwidth needed for training and inference, while NAND products provide the storage needed for datasets and AI models.

Importantly, Micron is not the market leader in DRAM or NAND products, but the company is gaining market share in both categories and its primary competitors, Samsung and SK Hynix, are losing market share. Particularly important, Micron gained 10 percentage points of market share in HBM over the past year.

Why is Morgan Stanley so bullish? The analysts argue the ongoing AI buildout has led to the most severe DRAM and NAND shortage in the last three decades. That shortage is driving prices higher and Micron is well-positioned to benefit. Indeed, Wall Street expects Micron’s earnings to grow at 48% annually over the next three years, which makes the current valuation of 28 times earnings look quite cheap.