A self-reinforcing cycle is accelerating across the artificial intelligence sector.
Smarter AI models require increasingly huge amounts of training compute and the rising consumer and enterprise adoption of agentic AI is driving a parallel surge in inference demand.
Compute capacity, not model design, has become the primary bottleneck constraining how fast the industry can scale, and that pressure is already visible in the market – GPU prices have increased by at least 25% year-to-date.
In their latest research report, Piper Sandler analysts led by David O’Connor shared their view on a group of compute and semiconductor stocks to evaluate this shifting market structure.
For them, industry heavyweights Nvidia (NVDA) and Santa Clara-headquartered Advanced Micro Devices (AMD) remain the top sector picks; the firm rates both at Overweight.
Nvidia: the market leader
O’Connor stamped a $300 price target on Nvidia stock, implying about a 34% upside from current levels.
He sees the titan as the outright AI compute market leader, commanding an impressive 80% market share.
In his research note, the analyst touted NVDA’s pace of product innovation as a significant differentiator.
Despite this dominance, Nvidia shares have lagged some of its peers in recent months – weakness that O’Connor doesn’t treat as a red flag but as a temporary mispricing only.
Piper Sandler flagged large-scale, gigawatt enterprise compute deals as the key near-term catalyst.
According to its analysts, NVDA currently trades at about 14x forward earnings, near the low end of the 14x to 22x range they assigned to the broader compute group, reinforcing the case that it remains inexpensive to own relative to its growth trajectory.
Nvidia’s management recently guided for $108 billion in sales for its current financial quarter, with a rather impressive 74% gross margin.
AMD: closing the gap
AMD stock secured a $600 price objective, the highest in Piper Sandler’s semiconductor coverage group, representing about a 15% implied upside.
The bull case centers on agentic artificial intelligence accelerating demand for CPU server chips.
As AI systems increasingly plan, orchestrate, and execute multi-step tasks rather than generating single responses only, that added overhead falls on general-purpose processors, not just GPUs – a segment where AMD is taking market share from legacy incumbent Intel.
AMD is also logging inference share gains against Nvidia through its Helios GPU line.
While the firm has secured major anchor customers for these chips, specifically OpenAI, Meta, and Anthropic, O’Connor indicated he wants to see Advanced Micro Devices broaden its core customer base before the company can realistically land its own gigawatt-scale enterprise compute deals.
AMD has guided for revenue of about $13 billion in its current financial quarter, which would represent a 41% growth versus the same quarter last year.
The post Piper Sandler names 2 stocks set to benefit from rising compute demand appeared first on Invezz