When investors think about AI, NVIDIA and other chipmakers usually come to mind first. But the latest market moves suggest the opportunity may be spreading further down the AI supply chain.
While some chip stocks have faced selling pressure, several AI infrastructure companies have moved sharply higher. Lumentum gained 11.04%, Bloom Energy rose 9.63%, Nebius climbed 7.73%, and CoreWeave jumped 11.72% in the market update.
These moves highlight an important question: Are investors starting to look beyond AI chips and focus on the infrastructure needed to make AI work at scale?
🔹 Lumentum — AI needs faster connections
AI data centres require enormous amounts of data to move between processors and servers. Optical networking technology helps meet that demand by enabling faster, more efficient data transmission. As AI infrastructure expands, connectivity could become an increasingly important part of the investment story.
🔹 Bloom Energy — AI cannot run without power
Building more data centres means finding reliable electricity to support their operations. Bloom Energy focuses on fuel-cell technology, offering an alternative approach to on-site power generation. With electricity availability becoming a key consideration for data-centre developers, energy infrastructure deserves attention alongside computing hardware.
🔹 Nebius — The demand for computing capacity
Nebius focuses on AI infrastructure and cloud computing. Its reported selection as Palantir’s preferred sovereign AI partner adds another development for investors to watch. The broader opportunity is clear: companies developing AI applications need access to computing capacity, but the challenge is turning demand into sustainable revenue and profits.
🔹 CoreWeave — Strong demand, but watch the cash flow
CoreWeave provides cloud infrastructure designed for demanding workloads, including AI training and inference. Its growth reflects the need for specialised computing capacity. However, rapid expansion requires substantial investment, and the company’s reliance on customer prepayments is worth monitoring. Revenue growth is encouraging, but investors should also examine capital expenditure, debt, cash flow and customer commitments.
What does this mean for investors?
I think the more interesting question is no longer simply which company will sell the most AI chips. It is which businesses can solve the bottlenecks preventing AI infrastructure from expanding.
Three areas stand out:
⚡ Power: Can energy suppliers keep up with growing data-centre demand?
🔌 Connectivity: Can optical networking companies benefit as data moves faster between increasingly powerful systems?
🖥️ Computing capacity: Can AI cloud providers expand quickly while maintaining healthy financial positions?
However, a rising share price does not automatically make a company a good investment. Lumentum’s reported 165.48% year-to-date gain, for example, shows how quickly expectations can rise. Even companies benefiting from genuine demand can experience sharp corrections if valuations get ahead of earnings.
My approach would be to look beyond the headlines and compare revenue growth, order backlogs, margins, capital expenditure and free cash flow. The strongest businesses will need to demonstrate that demand is translating into durable financial performance.
The next AI opportunity may be hiding in the infrastructure rather than the headline technology. The challenge is identifying which companies can turn that opportunity into lasting profits.
Which area would you watch most closely over the next 12 months: power, optical networking or AI computing infrastructure?
Disclaimer: Not financial advice. Do your own research before investing.
Comments