Arm (ARM 3.71%), the world's largest mobile chip designer, posted its latest earnings report on Nov. 5. For the second quarter of fiscal 2026, which ended Sept. 30, its revenue rose 34% year over year to $1.14 billion and exceeded analysts estimates by $80 million. Its adjusted earnings per share (EPS) rose 30% to $0.39 and cleared the consensus forecast by $0.06.
For the third quarter, Arm expects revenue to rise 25% year over year as adjusted earnings per share grows 5%. That outlook is stable, but Arm expects its near-term earnings growth to be throttled by its development of new chips and the expansion of its first-party chipmaking business.
That plan to sell its own first-party chips marks a major deviation from the company's traditional business model of licensing its chip designs to other chipmakers. Let's see why Arm wants to produce its own chips -- and why that shift might matter more than its recent earnings beat.
Image source: Getty Images.
Understanding Arm's business model
Arm's chip designs are currently used in about 99% of the world's smartphones. It conquered that market by designing smaller and more power-efficient chips than the x86 CPUs that Intel and AMD produced for PCs and servers. Instead of prioritizing raw processing power, Arm designed chips that consumed less energy -- making them ideal for mobile devices, wearables, cars, and Internet-of-Things (IoT) products.
NASDAQ: ARM
Key Data Points
Arm licensed those chip designs to top chipmakers like Qualcomm, MediaTek, and Apple. Most of those chipmakers were fabless ones that customized their chip designs and outsourced the manufacturing to leading foundries like Taiwan Semiconductor Manufacturing (TSMC).
A handful of integrated device manufacturers (IDMs) like Samsung, STMicroelectronics, and Texas Instruments manufactured their Arm chips at their own first-party foundries. All of these chipmakers paid Arm royalties and licensing fees to use its designs -- and that business model generates high-margin revenues.
Most of Arm's recent growth was driven by the market's robust demand for its AI-optimized Armv9 designs across the smartphone, cloud, and auto markets. These top-tier chip designs generate much higher royalties and licensing fees than its non-AI chip designs.
For many investors, Arm's diverse customer base and high-margin licensing model made it more appealing than more capital-intensive fabless chipmakers and IDMs. But in early 2025, Arm announced plans to become a fabless chipmaker by producing its own first-party chips.
Why is Arm transforming into a chipmaker?
That decision stunned many investors, since it could compress Arm's margins and drive it to compete against some of its top customers. However, Arm doesn't plan to produce any consumer-facing mobile system on chips like Qualcomm, MediaTek, or Apple. It's only producing server-class AI accelerators for hyperscale data center customers.
Arm will initially launch those chips as reference designs. Its data center customers can then use those designs to co-develop their own AI accelerators with Arm. They can either license those chip designs from Arm and handle the manufacturing on their own, or they can buy the finished chips from Arm -- which will outsource the actual production to TSMC.
Meta signed up as Arm's first major chip customer earlier this year. Other cloud giants like Amazon, Microsoft, Alphabet's Google, and Oracle could follow that lead as they expand their AI infrastructure. Arm's growth in data centers could hurt Intel, which still dominates more than 80% of the server market with its Xeon CPUs.
Arm's ambitious plans for the data center market could still impact Qualcomm and MediaTek, which are both expanding into servers with their latest AI-oriented chips. Since Arm doesn't need to pay its own royalties and licensing fees, it can likely afford to sell its AI accelerator chips at lower prices than its own chipmaking customers. But Qualcomm and MediaTek still generate most of their revenues from smartphones -- and Arm doesn't plan to expand its first-party chips into that saturated market yet.
But is it the right time to buy Arm's stock?
Arm's expansion of its first-party data center chip makes strategic sense, since the company can reduce dependence on the maturing smartphone market and give it a firm foothold in the AI market. The move could squeeze its near-term profits, but could generate fresh catalysts for the company's long-term growth.
From fiscal 2025 to fiscal 2028, analysts expect Arm's revenue and EPS to grow 20% and 34%, respectively, as sales of AI chips accelerate across multiple markets. However, its stock already trades at 121 times next year's earnings and 29 times next year's sales -- so the stock's upside potential could be limited in this frothy market. It might be worth nibbling on as a speculative AI chip play, but investors should probably wait for a pullback before accumulating more shares.
