Data center and AI demand was a little more common in earnings filings this quarter. Not dramatically more common, and not confined to chipmakers. The phrase reached into industrial equipment, infrastructure, security, energy and software spending.
In the August 19 to September 9 filing window, 40 companies flagged the driver, equal to 28.6% of filers. In the same-length window one quarter earlier, 48 companies did so, but that represented 24.4% of filers. On a share basis, the topic moved up about 1.17 times quarter over quarter. That is a tick, not a takeover.
The most direct version came from Marvell, where AI demand was attached to a broad product lineup rather than a single semiconductor niche.
"Strong revenue growth from our data center market was driven by AI-related demand for a broad range of our products, including electro-optics, custom, storage, and switching."
Marvell Technology, Inc. / 10-Q / August 28, 2026
Marvell's wording puts the emphasis on breadth. AI demand is reaching across the data center stack, at least in this company's account of the quarter.
BBCP tied the same theme to ordinary commercial activity, including pricing and project volume.
"The increase in revenue was driven by organic volume growth from growing commercial project demand including data center activity, infrastructure projects, and pricing improvements."
BBCP / 10-Q / September 3, 2026
That is the less glamorous but useful translation: data center construction is showing up as work for companies that sell into commercial projects, not just as revenue for the companies building processors and networking gear.
Methode Electronics described the demand through its industrial segment, alongside lighting products and customer recoveries.
"The increase was primarily driven by higher sales volume and mix in the Industrial segment due to organic growth in the data center business and an increase in demand for on-highway and off-highway lighting products (including customer recoveries)."
Methode Electronics, Inc. / 10-Q / September 2, 2026
The quote is upbeat, but it also shows why the theme needs some care. A filing can mention data centers as one contributor inside a wider industrial mix. The label does not mean every dollar of growth is an AI dollar.
Other companies used the same driver to describe constraints rather than sales. OSI Systems pointed to tighter component availability and higher costs.
"We have experienced tighter supply conditions and increased costs for certain memory associated and semiconductor components, reflecting a broader global imbalance between supply and demand for memory used in data center and AI related infrastructure."
OSI Systems, Inc. / 10-K / August 21, 2026
Napco Security described a similar risk in more conditional terms, saying AI infrastructure and data center expansion could affect its supply chain and operating results. That is a warning about exposure, not a report that the damage has already arrived.
"The decrease in information technology related expenses was primarily driven by a decrease in data center costs due to a reduction in cloud infrastructure capacity to align with current demand."
([SLQT](https://jodie.ai/t/SLQT) / 10-K / August 25, 2026)
SLQT supplies the useful counterexample. Data center demand can also appear as a reason to reduce cloud capacity and information technology spending. The topic's rising share in filings therefore says more companies are discussing the same force, not that they are describing one uniform outcome.
The approximate tone tally leaned positive, with 15 tagged mentions reading positive and two negative. The quotes explain the split: Marvell, BBCP and Methode describe demand feeding revenue, while OSI describes scarce components and SLQT describes lower cloud capacity. AI and data center language is broadening across industries, but the business effect still depends on where a company sits in the chain.
This is a descriptive reading of SEC filings via jodie's analytics, not investment advice.