Nvidia and Micron: AI Slowdown Will Punish One AI Stock More Than the Other, Says Investor

Calls to slow the pace of frontier AI development put the semiconductor sector back on edge on Monday. Anthropic CEO Dario Amodei recently argued that the industry should “pace the frontier” to give companies more time for safety testing, oversight and governance, with OpenAI CEO Sam Altman and other AI leaders generally supportive of the idea. The market reaction was immediate, with several semiconductor stocks selling off sharply as investors questioned whether slower model development could eventually mean less spending on AI infrastructure.

But slowing the development of ever-more-powerful models does not necessarily mean slowing AI infrastructure demand. The bigger shift may be from training the next frontier model toward running AI at enormous scale. As inference costs fall, companies can afford to use AI for more tasks, while agentic systems can generate huge numbers of tokens through repeated reasoning, planning and execution. That changes what matters in the semiconductor supply chain.

An investor with the moniker Summit Research (SR) argues that this distinction is particularly important when comparing Nvidia (NASDAQ:NVDA) and Micron (NASDAQ:MU). Both benefit from AI spending, but their exposure is different, leaving Nvidia more vulnerable if the industry places less emphasis on rapid frontier upgrades.

For Nvidia, the problem is not that demand for compute disappears. Inference should remain a huge source of demand. The issue is that a slower frontier could reduce the urgency to buy Nvidia’s newest and most powerful accelerators. If AI labs take an extra year or two between major model advances, customers may have less reason to rush into each new generation of hardware.

That matters because Nvidia has built its business around a rapid upgrade cycle. Each generation promises substantial gains in performance and efficiency, allowing the company to sell customers more powerful systems and support premium pricing. If frontier progress slows, however, the monetization of that annual cadence could become less predictable. Customers may continue buying Nvidia hardware, but they could put greater emphasis on cost-effective inference rather than paying up for maximum performance.

There is also a potential margin issue. Nvidia has argued that technological leadership and major performance-per-dollar improvements are central to sustaining its gross margins. A shift toward cheaper, inference-focused systems could therefore create a less favourable product mix, particularly if customers become less willing to pay for the very latest hardware.

Micron has a different exposure. Its HBM is tied closely to Nvidia and other AI accelerators, so a slowdown in frontier upgrades would still create some risk. But Micron also supplies server DRAM and NAND storage, areas that can benefit directly from the sheer volume of AI workloads.

Agentic AI requires large amounts of memory and storage to handle expanding context windows, KV caches and persistent data. That means Micron can benefit even when the AI industry is running more workloads on existing or less cutting-edge accelerators. Its opportunity is therefore tied more closely to the number of tokens being processed than to how quickly the frontier model itself advances.

SR also points to a crucial supply constraint. Micron has indicated that its HBM capacity is already accounted for well into 2028, while conventional DRAM is also facing tight supply. If frontier demand softens, capacity could potentially be redirected toward other memory products needed for large-scale inference.

That gives Micron a wider cushion. Nvidia remains heavily exposed to the pace of its accelerator upgrade cycle, while Micron can participate across several layers of the AI memory hierarchy.

As such, SR has downgraded Nvidia’s rating to Hold (i.e., Neutral), while the investor maintained a Buy rating for Micron. (To watch Summit Research’s track record, click here)

On the other hand, the Street has a bullish stance for both names, giving the pair Strong Buy consensus ratings. At $324.30, NVDA’s average price target points toward 12-month returns of 54%. Micron’s average target stands at $1,563.93, a figure offering one-year upside of 69%. (See Nvidia stock forecast or Micron stock forecast)