AAPL

TSLA

AMZN

META

AMD

NVDA

PEP

COST

ADBE

GOOG

AMGN

HON

INTC

INTU

NFLX

ADP

SBUX

MRNA

Markets

AMD

The "Great Rotation" Out of Tech Is Fading. Here Are the Best Artificial Intelligence (AI) Growth Stocks Poised to Benefit.

April 25, 2026 — 02:05 pm EDT

Written by

Geoffrey Seiler for

The Motley Fool->

Key Points

- AMD has big opportunities with inference and agentic AI.

- Broadcom is set to see huge growth driven by custom AI chips.

- Micron is riding a power memory wave that shows no signs of slowing.

- 10 stocks we like better than Advanced Micro Devices ›

The so-called "Great Rotation" out of tech and artificial intelligence (AI) stocks appears to have been short-lived. Given that tech and growth stocks have helped power the market higher for much of the past two decades, this probably shouldn't be surprising. While I wouldn't completely write off value stocks, growth is still the place to be, and right now, the biggest growth is coming from AI.

Let's look at three AI stocks to buy right now.

Will AI create the world's first trillionaire? Our team just released a report on the one little-known company, called an "Indispensable Monopoly" providing the critical technology Nvidia and Intel both need. Continue »

AMD: An inference and agentic AI winner

While the first phase of AI was all about AI model training, the next phase of AI will be more about inference and AI agents. This is great news for Advanced Micro Devices (NASDAQ: AMD), which is uniquely set to benefit from these trends.

Long an afterthought in the data center graphics processing unit (GPU) market, the strides the company has made with its ROCm software and the chiplet design (which packs more memory into its chips) of its upcoming MI450 GPU help put the company in the inference conversation. Two big partnerships with OpenAI and Meta Platforms, meanwhile, should help drive growth.

The even more exciting area for AMD, though, is data center central processing units (CPUs). The company is the leader in this space, and as agentic AI rises, the CPU-to-GPU ratio in AI data centers is about to get much tighter. The data center CPU market is already tight, and AMD is set to deliver new CPUs designed specifically for agentic AI.

Between its GPU and CPU opportunities, this is a stock to own.

Image source: Getty Images.

Broadcom: Custom AI chips set to be a huge driver

As hyperscalers continue to look for ways to become less reliant on Nvidia, another company they are increasingly turning to is Broadcom (NASDAQ: AVGO). Broadcom is a leader in ASIC (application-specific integrated circuit) technology and helped Alphabet develop its highly successful tensor processing units (TPUs). With TPU growth booming, Broadcom is riding a powerful wave, both from Alphabet's internal use of the chips and from its growing acceptance by other large customers. This includes Anthropic, which has ordered $21 billion in chips from Broadcom to be delivered this year, and has recently extended its partnership with Alphabet and Broadcom for future TPU deliveries.

Meanwhile, other companies have also been turning to Broadcom to help them develop their own custom AI chips, including OpenAI. Broadcom has projected that it would sell $100 billion in AI chips alone in fiscal 2027, about 5 times the total AI revenue it produced last year. Meanwhile, it is also seeing robust growth from its data center networking business, where it is a leader in the space with its Tomahawk Ethernet solution.

With Broadcom set to see growth skyrocket, now is a time to add the stock.

Micron Technology: Riding the memory wave

While you can argue that Micron Technology (NASDAQ: MU) is "over-earning" in this current memory cycle, what you can't argue is that the company is riding a huge trend that isn't likely to end anytime soon. As one of the big three DRAM memory makers, along with Korean companies Samsung and SK Hynix, the company is benefiting from a DRAM shortage driven by AI, leading to strong revenue growth and robust gross margins.

The shortage is directly tied to GPU and AI chip demand, as these chips require packaging with a specialized form of DRAM called high-bandwidth memory (HBM) to optimize performance. As such, demand is tied directly to increasing AI compute power needs. On top of that, HBM requires upward of 3 times the wafer capacity of ordinary DRAM, helping add to the overall tightness of the DRAM market.

Given that the growing demand for HBM is directly tied to AI chip growth, as AI infrastructure booms, so should the DRAM market. Meanwhile, Micron and its Korean competitors are starting to lock up long-term HBM deals for the first time, which should help take out some of the business's cyclicality and raise the overall