Analyst Identified Four Key AI Infrastructure Firms
Seaport Research categorized Marvell, Semtech, ON Semiconductor, and SiTime as the "four horsemen" of AI hardware.
Updated on Sept. 23, 2026 in Semiconductors

Live Poll
Is now a good time to diversify your portfolio with lesser-known AI hardware stocks?
Analyst Jay Goldberg initiated coverage on four semiconductor companies he identified as the primary drivers of artificial intelligence infrastructure. He assigned buy ratings to Marvell Technology, Semtech, and ON Semiconductor while issuing a neutral rating for SiTime.
Why it matters
These firms provide the essential hardware for data center networking and synchronization required by the current industry-wide push for AI acceleration. The move highlights the market focus on specific components that bridge the gap between individual processors and large-scale AI clusters.
The analyst identified these firms for their roles in AI infrastructure, ranging from Semtech's networking connections for graphics processing units to SiTime's precision synchronization components. These specialized chips are designed to optimize data flow across massive AI data center architectures.
The players
Jay Goldberg
Seaport analyst who follows the semiconductor industry and hardware supply chains.
Marvell Technology
Chip designer and manufacturer with a significant focus on custom-silicon solutions for data centers.
Semtech
Company specializing in networking connectivity hardware that enables data transmission between AI processors.
ON Semiconductor
Semiconductor firm that manufactures chips for automotive and mobile AI, having expanded through its acquisition of Synaptics.
SiTime
Manufacturer of precision timing and synchronization components required for electronic system stability.
The details
The companies address distinct bottlenecks in AI infrastructure. Semtech facilitates high-speed networking between graphics processing units, while SiTime provides timing synchronization necessary for complex electronic signals. Marvell Technology differentiates itself through a custom-chip business, and ON Semiconductor focuses on real-world AI applications for smartphones and automotive markets, a strategy bolstered by its acquisition of Synaptics.
Timeline
Semtech acquired Sierra Wireless in 2023.
Analyst Jay Goldberg issued the note initiating coverage on September 22, 2026.
Stock prices for the four companies dropped at the market open on September 23, 2026.
The Tech Race
The analyst's framework prioritizes firms with specific intellectual property in the custom-chip market, which has become the primary battleground against competitors like Broadcom and Credo. By grouping these firms, the assessment highlights the shift toward vertical specialization in AI hardware.
Market observers should monitor the performance of these stocks as these companies attempt to scale their hardware for broader AI data center integration. While the analyst is optimistic about the custom-chip growth potential, investors will likely watch for future revenue disclosures to confirm the efficacy of these specific AI-focused strategies.
The takeaway
The sector's trajectory is increasingly defined by the ability to provide specialized, high-performance networking and timing components for AI hardware. Investors and engineers should track the revenue growth of Marvell's custom-chip division as a bellwether for the broader AI infrastructure market.
Further reading
For more on the hardware driving modern machine learning, visit Semiconductors.
Source note: This article includes information reported by Morningstar.
Live Poll
Is now a good time to diversify your portfolio with lesser-known AI hardware stocks?









