JPMorgan is sharpening its bets on the artificial intelligence hardware boom, singling out Nvidia, Broadcom and Micron as its preferred chip names heading into a pivotal earnings stretch. The call underscores how quickly the center of gravity in semiconductors has shifted toward companies that either power AI computation directly or supply the memory and networking needed to keep those systems fed.
As investors brace for fresh guidance on data center spending, these three stocks have already reacted, with shares of several U.S. semiconductor companies moving higher after JPMorgan highlighted them as top picks ahead of the next reporting wave. I see that move less as a short-term trade and more as a referendum on which parts of the chip stack Wall Street believes can sustain pricing power as AI demand broadens.
Nvidia’s AI dominance and valuation tension
Nvidia sits at the heart of this story, with a market valuation that is nearly three times higher than its closest American semiconductor rival and coverage of about 80% of the AI chip market. That kind of concentration gives Nvidia enormous leverage over the direction of AI infrastructure, from the GPUs inside Microsoft Azure clusters to the systems training frontier models at OpenAI and Meta. Visual industry rankings show Nvidia in an “unquestionably dominant position” atop the U.S. semiconductor industry, which helps explain why JPMorgan is comfortable putting the stock at the front of its chip roster before earnings.
That dominance is not just about market cap, it is rooted in Nvidia’s role at the center of the AI revolution, where its high performance graphics processing units are essential for training and running large language models and other compute hungry workloads. As enterprises race to deploy generative AI, the demand for Nvidia’s chips has skyrocketed, a dynamic that has already driven spectacular revenue growth and fueled debate about how long the cycle can last. Some market voices now warn of a potential Nvidia stock correction, arguing that expectations have become stretched, but JPMorgan’s decision to keep the company among its preferred chip names suggests it still sees earnings as the main arbiter of whether the valuation is justified.
Broadcom’s AI networking edge
If Nvidia is the face of AI compute, Broadcom has quietly become one of the biggest beneficiaries of the AI networking build out that makes those GPUs useful at scale. The company has emerged as a key supplier of custom silicon and high speed networking gear that ties together racks of accelerators inside hyperscale data centers. Analysts tracking the AI infrastructure wave highlight that Key Points Broadcom is positioned to capture around 20 billion dollars of non accelerator AI revenue, largely from this networking upgrade cycle, which helps explain why JPMorgan is comfortable elevating the stock alongside Nvidia.
In my view, Broadcom’s appeal in this trio is that it offers AI exposure without relying solely on GPU pricing or unit volumes. Its custom chips for cloud providers, combined with Ethernet and optical components, give it a diversified stream of AI related sales that can grow even if GPU supply tightens or customers experiment with alternative accelerators. That diversification, paired with a history of disciplined capital returns, makes Broadcom a natural candidate for investors who want to participate in the AI infrastructure boom but prefer a slightly different risk profile than a pure compute play.
Micron and the memory backbone of AI
Micron rounds out JPMorgan’s preferred chip basket as the memory specialist that stands to benefit from the sheer data intensity of AI workloads. Training and inference both require vast amounts of high bandwidth memory and fast storage, and Micron’s portfolio of DRAM and NAND gives it direct leverage to that trend. When JPMorgan flagged Nvidia, Broadcom, Micron as top chip picks ahead of earnings, it was effectively endorsing the idea that memory pricing and volumes are set to improve as AI servers consume more bits per system.
Memory has historically been one of the most cyclical parts of the semiconductor industry, with brutal downturns when supply overshoots demand. The AI era does not erase that risk, but it does change the slope of the demand curve, since each new generation of accelerator tends to require more high bandwidth memory per GPU and more capacity per node. If Micron can manage its capital spending and product mix to align with that structurally higher baseline, the company could see a longer stretch of favorable pricing than in past cycles, which is likely a key reason JPMorgan is willing to lean into the stock before it reports.
ASML and the broader equipment backdrop
JPMorgan’s enthusiasm for Nvidia, Broadcom and Micron sits within a wider call on the semiconductor capital equipment ecosystem, where it has also turned more constructive. The bank recently raised its target on ASML and added the stock to its Analyst Focus List, signaling confidence that the company’s extreme ultraviolet lithography tools will remain a bottleneck technology for advanced chip production. In its note, JPMorgan described the semiconductor capital equipment group as the most attractive segment of the sector and argued that ASML’s systems would be the main source of upside as foundries ramp leading edge capacity, a view reflected in the decision to place the shares on the Analyst Focus List.
For investors weighing JPMorgan’s chip preferences, that equipment call matters because it reinforces the idea that the AI build out is still in an early investment phase. Foundries and integrated device manufacturers cannot deliver the GPUs, networking chips and memory that Nvidia, Broadcom and Micron sell without first buying advanced tools from companies like ASML. By backing both the end product leaders and the critical equipment supplier, JPMorgan is effectively expressing confidence in the entire AI manufacturing chain, from lithography to packaged accelerators.
AI demand, competition and policy risk
Underpinning all of these stock calls is a simple reality, AI workloads are driving a structural surge in demand for high performance graphics processing units and related silicon. Market research on the graphics processing unit market notes that companies such as NVIDIA and AMD are at the forefront of developing high performance GPUs to meet these needs, as customers continue to prioritize computational speed and precision in data centers, gaming rigs and professional workstations. That report argues that this surge in demand will persist as more industries adopt AI, with NVIDIA, AMD leading the charge, which helps explain why JPMorgan is comfortable centering its chip thesis on AI exposed names.
At the same time, competition and policy risk are intensifying around Nvidia in particular. AMD, now the third largest U.S. chipmaker by market cap, has struck a multibillion dollar deal with OpenAI to supply chips for its data centers, a move that directly targets Nvidia’s share of AI accelerator sockets. Policy makers are also scrutinizing advanced chip exports and global supply chains, and market commentary notes that Highlights Nvidia remains central in semiconductor discussions as governments debate rules on advanced computing and international technology distribution. Those cross currents mean that while JPMorgan’s trio of Nvidia, Broadcom and Micron may be well positioned for the next leg of AI driven growth, investors still need to watch how competitive dynamics and regulation evolve as the earnings season unfolds.
More from Morning Overview