This article is written in conjunction with Kevin Hein of Tirias Research, who is attending the trial proceedings.

Arm’s move into chips is changing its relationship with licensees and raising larger questions about neutrality, technological sovereignty and the future of the computing ecosystem.

Arm has sometimes been described as the “Switzerland” of the semiconductor industry. The analogy was never perfect, but the idea was easy to understand. Arm developed processor architectures and technology that other companies used to build chips, while largely staying out of their way.

Apple could build processors for the iPhone. Qualcomm could build Snapdragon. Nvidia could build processors for automotive and data centers. MediaTek could build smartphone chips. Amazon, Google and Microsoft could develop their own silicon.

They competed fiercely with one another, but Arm provided common ground underneath much of that competition. That model helped Arm become one of the most important technology companies in the world. It also makes what is unfolding in the Qualcomm-Arm trial in Wilmington much more significant than a fight over licensing rates. What is emerging from the testimony is what happens when Arm is no longer content to remain Switzerland.

Arm Is Becoming A Chip Company

Arm CEO Rene Haas made the company’s changing direction unusually clear during testimony Wednesday.

Asked whether roughly two-thirds of Arm’s projected revenue would come from chips by 2030, Haas said two-thirds of revenue would be from chips and added, “we are adding to our business a chip business.” He rejected the suggestion that this would necessarily hurt Arm’s partners, pointing to support Arm has received from companies including Nvidia, AWS, Google and Microsoft.

Arm itself now describes the move into production silicon as a “major strategic evolution” beyond IP licensing and compute subsystems. There is nothing inherently wrong with that evolution. Arm has every right to look for ways to grow its business and capture more of the value being created around its technology.

Haas also offered a reasonable explanation for why Arm believes the economics surrounding its architecture should change. The semiconductor market of 2026 is dramatically different from the one in 2013 when Qualcomm and Arm negotiated their current architecture agreement. Arm’s presence in data centers was tiny. Its automotive opportunity was much smaller. Smartphone processors were less expensive and contained far fewer CPU cores.

Haas testified that a smartphone processor that might have sold for $20 to $30 in 2013 can sell for around $220 today. Server processors can contain 100 or 200 cores and sell for thousands of dollars. Arm believes the value it brings to companies building these products has increased along with those markets.

That is a legitimate business argument, but there is another side to it. The more Arm participates directly in the markets its customers serve, the more complicated its relationship with those customers becomes.

The Nvidia Question Returns

One of the most interesting exchanges Wednesday concerned Nvidia’s failed attempt to acquire Arm.

Nvidia agreed to buy Arm from SoftBank in 2020 for $40 billion, but regulators objected. The U.S. Federal Trade Commission argued that allowing a major chip company to control Arm could give it the ability and incentive to disadvantage competitors that also depended on Arm technology. The FTC specifically cited Arm’s neutral licensing model and its reputation as the “Switzerland” of the semiconductor industry.

Nvidia and SoftBank attempted to address exactly that concern. When the acquisition was announced, they promised Arm would continue its open licensing model while maintaining the “global customer neutrality” that had been fundamental to its success. The acquisition was eventually abandoned in 2022 after regulatory opposition in multiple countries.

That history became relevant again in Wilmington. Haas was asked whether one concern surrounding the Nvidia acquisition was that a chipmaker owning the Arm instruction set architecture could advantage itself against other chipmakers. He answered yes.

The Nvidia acquisition never happened. But several years later, the underlying question has returned from a different direction. Nvidia did not become the owner of Arm. Arm is becoming a chip company itself. That doesn’t mean Arm will unfairly disadvantage Qualcomm or anyone else. It does mean the relationship is changing.

A company that once primarily supplied technology to chipmakers increasingly wants to participate in more of the value created from that technology.

Arm And Qualcomm Still Need Each Other

One of the more revealing parts of Haas’s testimony was that he did not portray Qualcomm as an enemy Arm wants to eliminate. Quite the opposite. Haas said Arm wants Qualcomm to succeed in PCs, automotive and data centers. He described Intel and AMD as common competitors and later made the relationship even clearer.

Qualcomm depends on Arm, Haas said, and Arm depends on Qualcomm. Qualcomm invests billions of dollars developing products around Arm technology. Arm benefits when those products succeed. Haas described x86, RISC-V and other architectures as the real architectural competition and said logic suggests the two companies should work out their differences.

That gets to the strange heart of this trial. Arm and Qualcomm need each other while simultaneously moving into more of the same markets. And the stakes become higher each time Arm advances its architecture.

Qualcomm wants access to Arm v10 under economics it believes allows it to remain competitive. Arm believes Qualcomm’s existing agreement reflects a world that existed more than a decade ago and does not adequately represent the value of Arm technology today.

Haas was direct about who should ultimately determine the value of that technology. “It’s our product,” he testified. “Of course we should decide. If we don’t decide who should decide?” That is difficult to argue with as a basic statement of ownership.

Arm Is Not Just Another Supplier

Switching architectures is not the same as changing memory suppliers or moving a chip between two manufacturing processes. An architecture requires an ecosystem around it: operating systems, compilers, development tools, applications, engineering expertise and decades of software, much of which is at the expense of its partners, including chip developers like Qualcomm. The billions of dollars invested by that ecosystem is part of what makes Arm so valuable. It also makes Arm strategically important far beyond Qualcomm.

Arm remains a British company headquartered in Cambridge, but its controlling shareholder is Japan’s SoftBank. SoftBank acquired Arm in 2016 for approximately $31 billion and promised to maintain its Cambridge headquarters, preserve its business model and invest heavily in its UK workforce. Arm returned to the public market in 2023, but SoftBank still owns 86.4% of the company and Arm remains a controlled company according to Arm’s most recent SEC filing.

Then there is China. China represents an important semiconductor market for Arm, but Arm reaches it through an unusual Arm China structure. Arm says neither it nor SoftBank controls Arm China’s operations. Arm also warns investors that tensions between China, the United States and the United Kingdom over trade and national security can affect its ability to license technology to Chinese companies. That turns architecture into something larger than a commercial licensing decision.

From Corporate Strategy To National Strategy

The semiconductor industry is increasingly divided by national priorities. The United States wants leadership in AI and advanced computing. China wants greater semiconductor independence. Europe wants greater technological sovereignty. Japan is investing heavily in rebuilding its semiconductor industry.

Arm sits across those boundaries. Its architecture has succeeded precisely because companies around the world have been able to build around it. The more strategically important Arm becomes and the more value Arm attempts to capture itself, the more incentive companies and governments have to consider alternatives.

That is where RISC-V becomes important. RISC-V does not replace the Arm ecosystem tomorrow. Qualcomm’s own testimony this week highlighted how much work and is required to build the software, tools and broader ecosystem necessary to support another architecture across major computing markets. Qualcomm has huge investments in building an Arm ecosystem for smartphones, PCs, wearables, and other markets. But RISC-V offers something strategically different: an open instruction set architecture that is not commercially controlled by one semiconductor technology supplier.

For Qualcomm, investing in RISC-V provides another path for future products and potentially reduces its dependence on Arm, but it requires that same ecosystem investment that it has already spent on Arm. For countries worried about access to foundational computing technology, the same characteristic becomes even more important. The irony is that the more valuable Arm becomes, the more valuable an independent alternative to Arm becomes as well.

What Comes After Switzerland?

The trial in Wilmington is exposing a transition that extends well beyond the dispute between Arm and Qualcomm. Arm spent decades creating an architecture and working with its partners to develop an ecosystem that allowed other companies to compete on top of the Arm architecture. That model helped Arm technology spread from consumer and embedded applications to smartphones, PCs, automobiles, cloud data centers and now AI infrastructure. Arm now wants to capture more of the enormous value that ecosystem creates.

Qualcomm wants assurance that its dependence on and investment in that ecosystem will not leave it facing dramatically different economics whenever the architecture advances. Both positions make sense and the tension comes from trying to reconcile them.

Arm can be a technology supplier. Arm can also become a much larger chip company. But doing both changes the relationship with companies that built their businesses around Arm’s historic role and invested in the development of the ecosystem. The Nvidia acquisition raised the question several years ago: What happens when a chip company controls an architecture its competitors depend on? Regulators never had to find out.

Now the industry is confronting a variation of the same question. What happens when the architecture company becomes the chip company? What happens when Arm stops being Switzerland?

Tirias Research tracks and consults for companies throughout the electronics ecosystem from semiconductors to systems and sensors to the cloud. Members of the Tirias Research team have consulted for IBM, Nvidia, Qualcomm, AMD and other companies throughout the mobile, data center, AI and Quantum ecosystems.