Skip to the letters
Machine Learning

Applied Materials Revenue Climbs 53% as AI Chip Equipment Demand Holds

from Mike Wheatley 789 reads

Applied Materials reported $10.41 billion in quarterly revenue, up 53%, and guided above forecasts as AI-driven chipmaking equipment demand keeps climbing.

Applied Materials Revenue Climbs 53% as AI Chip Equipment Demand Holds

Applied Materials doesn't make chips, but its numbers tell you a lot about who does. The company is the world's largest supplier of semiconductor manufacturing equipment, and when it reports strong results, it usually means chipmakers are spending on new capacity. This quarter, that spending looks robust in the best sense of the word: adjusted earnings and revenue both beat Wall Street estimates, and the company guided higher for the current quarter.

The headline numbers

For the fourth quarter, Applied reported adjusted earnings of $4.16 per share on revenue of $10.41 billion, up 53% from a year earlier. Analysts had modeled $4.06 per share on sales of $10.3 billion, so the company cleared both bars. Net income rose to $2.71 billion from $1.89 billion in the same period a year ago.

The full fiscal year was the bigger story. Chief Executive Gary Dickerson described fiscal 2026 as a record year, with full-year sales rising 21% to $34.31 billion. He also said the semiconductor systems business is on track to grow more than 40% in calendar 2026. Those are the kinds of figures that suggest the spending cycle has more room to run than skeptics expected a year ago.

Dickerson tied the outlook directly to AI. "AI continues to fuel strong incremental demand for semiconductors and Applied has the most-enabling portfolio to advance the AI computing roadmap, putting us in a great position for multiyear revenue and profit growth," he said.

Where the growth came from

The semiconductor systems unit is where the real acceleration shows up. Revenue in that segment jumped from $4.88 billion a year earlier to $7.99 billion. That business makes the deposition, etch, and other process tools that chipmakers use to build advanced logic and memory, and it's the segment most directly exposed to fab construction and upgrades.

The applied global services segment, which covers maintenance, upgrades, and support for installed equipment, grew more modestly, from $1.5 billion to $1.89 billion. That's still solid, and it matters for a different reason. Service revenue tends to follow the installed base, so steady growth there suggests the tools already in customers' fabs are running hard. Here's the thing: equipment sales spike when customers build, while services revenue reflects how intensively they use what they've already bought. Both rising at once is a healthy sign.

Guidance and what it implies

For the current quarter, Applied expects earnings of $4.35 per share at the midpoint of its guidance range, with revenue of about $11.1 billion. Analysts had been modeling $4.24 per share on sales of $10.85 billion. A guidance beat is often more useful to watch than a backward-looking one, because it reflects what the company can see in customer orders and fab schedules.

The gap between the numbers and the consensus is moderate, not dramatic. That's typical of a supplier with long, visible order books. Equipment makers usually know a lot about customer plans months ahead, so when they raise guidance, it tends to reflect confirmed commitments rather than optimism alone.

Stock reaction and the Nvidia distraction

Applied's shares were under pressure earlier in the session after Nvidia Corp. announced that CEO Jensen Huang was stepping down. Nvidia is the dominant supplier of AI accelerators, so investors reasonably asked whether turmoil there could ripple through the broader chip supply chain. That worry is worth keeping in view, since equipment orders ultimately depend on how much AI hardware the market keeps buying.

The results did the work of reversing the early slide. Applied's stock gained more than 2.5% in late trading, and it's up 48% so far this year. The market seems to be treating the Nvidia news as a leadership story rather than a demand story, at least for now. Whether that holds will depend on whether Nvidia's own orders and roadmap stay on track.

The EPIC Center bet

During the quarter, Applied also committed $5 billion to a new research and development facility in Silicon Valley called the EPIC Center. The idea is a shared R&D hub where chipmakers and equipment suppliers can develop new process technologies together, rather than each company building its own experimental lines from scratch.

Initial partners include Intel Corp., which is working with Applied and others on next-generation transistors, interconnects, and advanced packaging. Other founding participants include Kioxia Corp., BE Semiconductor Industries N.V., and the UCLA Samueli School of Engineering. The academic and multi-vendor mix is notable. Process development has become so expensive and interdependent that collaborative labs are an increasingly common way to share the cost and speed up learning.

For developers and infrastructure buyers, the practical question is timing. Advanced packaging and new transistor structures are what determine how much compute fits in a given power and space budget, and those gains reach data centers only after years of process work. A facility like EPIC is a bet that the next wave of AI performance depends as much on manufacturing know-how as on chip design.

What to watch next

The key variable is whether chipmakers keep converting AI demand into factory spending. Applied's guidance suggests they are for now, but equipment cycles have a history of turning quickly, and a single slowdown in one customer category can change the picture. Watch the semiconductor systems growth rate against the 40% target, and watch whether services keeps pace as the installed base expands.

My read is that the quarter confirms a strong near-term cycle rather than a new baseline. The companies building the most advanced AI silicon need the most advanced tools, and Applied is positioned close to that demand. The open question is how long the build-out runs at this pace, and that's the thing the next few quarters will answer.

Source: Mike Wheatley · siliconangle.com

Same correspondent