Arm Holdings: how customer relationships are accelerating an AI-driven royalty engine

Arm licenses CPU architectures and related IP to chipmakers and OEMs and collects fees through long-term architecture licenses and per-unit royalties; its monetization depends on pervasive deployment across mobile, edge, cloud and now AI server stacks. Recent FY2026 reporting and market coverage show Arm moving from mobile ubiquity toward strategic, long-dated partnerships with hyperscalers and OEMs—shifting the company toward higher-margin, recurring royalty upside tied to AI deployments. For an investor, the question is whether those large, multi-year deals increase revenue visibility or concentrate exposure to a handful of hyperscalers. Learn more at https://nullexposure.com/.

Why the relationship map matters for valuation and risk

Arm’s business model is inherently licensing-first: the company signs long-term contracts and captures value through design licenses and per-chip royalties, rather than selling physical silicon. That contracting posture produces a predictable revenue stream when license volumes scale, but it also concentrates economic sensitivity in a relatively small set of strategic partners as the company pursues AI-optimized architectures.

Concentration: News coverage in FY2026 highlights growing ties to Apple and major cloud providers, implying revenue concentration toward a handful of customers.
Criticality: Big-tech adoption turns Arm’s architecture into a critical input for AI servers and edge devices—raising strategic leverage but also political and supply-chain scrutiny.
Maturity and posture: Arm’s transition from mobile IP leader to AI infrastructure enabler reflects a mature IP portfolio being redeployed via long-term licensing and custom silicon collaboration.

Read more background and scenario analysis at https://nullexposure.com/.

The customer roster and what each relationship implies

Apple

Apple has extended a long-term licensing relationship with Arm that reportedly runs beyond 2040, signaling one of the most durable consumer-facing revenue anchors for Arm and underwriting long-run royalties tied to Apple silicon. According to Finviz and SimplyWall.st coverage in FY2026, Apple’s renewed commitment sharpens Arm’s consumer-device moat and reduces short-term churn risk.

Amazon Web Services (AWS)

AWS has used Arm-based processors in its data centers since 2018 and scaled to Arm parts capable of modern AI workloads by 2023, making AWS a strategic cloud partner for server-grade Arm adoption. A FY2026 Finviz article summarizes AWS’s multi-year adoption cycle and ongoing role in expanding Arm into cloud AI infrastructure.

Amazon (retail/other)

Beyond AWS, reporting in early 2026 notes Amazon’s broader product and internal compute initiatives building on Arm architectures for new AI and server processors, indicating multiple commercial touchpoints with Arm across the company. SimplyWall.st and SahmCapital write-ups in FY2026 highlight Amazon among the set of cloud and device customers building on Arm IP.

Google / Alphabet

Google tapped Arm for the architecture behind its Axiom processors designed for AI data centers, underlining hyperscaler-level design wins that convert Arm IP into custom server implementations. Coverage from Finviz and SahmCapital in FY2026 reports Google’s adoption as part of its vertically integrated AI hardware strategy.

OpenAI

Industry reporting in FY2026 connects Arm to an AI XPU ASIC program developed in partnership with SoftBank and Broadcom for OpenAI, pointing to higher-concentration, high-value engineering programs that could create meaningful royalty and design-fee upside if production scales. Finviz and SimplyWall.st note Arm’s role in that program.

Qualcomm

Qualcomm remains both a partner and a competitor: its mobile processors are built on Arm cores even as it competes in adjacent compute markets. Coverage in The Globe and Mail and TradingView in FY2026 highlights Qualcomm’s unique position as a large licensee and strategic peer in mobile and edge compute.

MediaTek

Reports in FY2026 list MediaTek among chipmakers collaborating on Arm-based system-on-chips for laptops and other devices, illustrating Arm’s continued penetration of client computing beyond phones. TradingKey coverage references upcoming Nvidia–MediaTek Arm-based laptop SoCs.

Nvidia

Nvidia shows up in FY2026 reporting as a collaborator in Arm-based PC and AI system efforts, particularly through joint SoC initiatives, underscoring cross-company engineering work linking Arm IP with discrete GPU and accelerator ecosystems. TradingKey and related coverage discuss Arm-based Nvidia–MediaTek SoCs for PCs.

Taiwan Semiconductor Manufacturing Company (TSMC)

Arm’s architecture is implemented by foundries like TSMC, which construct chips using Arm design instructions—TSMC’s central role in turning Arm IP into silicon makes the foundry relationship operationally critical. The BBC’s technology reporting in FY2026 reiterates TSMC as a principal manufacturer for Arm-based chips.

Arm China

News in FY2026 flags material revenue exposure (reported around 20–25%) to Arm China, an entity not fully controlled by Arm Holdings, creating geopolitical and concentration risk that could influence near-term revenue reliability. TradingKey’s FY2026 coverage highlights this exposure and the attendant geopolitical risk.

SoftBank

SoftBank is cited as a partner in the AI XPU program and is also a sizeable related-party given its historical ownership links; analysts in FY2026 raised questions about SoftBank’s role in Arm revenue flows and the potential for circular financing concerns. TradingKey commentary raises those governance and revenue-quality questions.

Broadcom

Broadcom is part of the reported AI XPU program alongside SoftBank and Arm for OpenAI, positioning Broadcom as a strategic collaborator in high-performance AI ASICs where Arm supplies the architecture. SimplyWall.st and Finviz pieces in FY2026 reference Broadcom’s technical and commercial role.

Samsung

Samsung is named among large manufacturers that use Arm’s design instructions to build their own processors, anchoring Arm’s relevance across multiple major OEMs and foundry relationships. The BBC wrote in FY2026 about Samsung’s use of Arm technology in chip construction.

What this roster means for investors

Upside: Major design wins with Apple, Google, Meta, AWS and OpenAI position Arm to capture accelerating royalties as AI workloads scale across cloud and edge. These partnerships support consensus growth assumptions if execution and volume scale as described.
Concentration & geopolitical risk: Significant exposure to Arm China and to a small number of hyperscalers increases revenue concentration and geopolitical sensitivity, which investors must price explicitly into multiples or scenario analyses.
Revenue quality and governance: Interactions with SoftBank and large custom programs raise questions about the long-term profile of licensing versus bespoke engineering revenue; analysts in FY2026 flagged circular-financing and revenue-sustainability concerns.

Bottom line and next steps

Arm has converted decades of mobile dominance into a strategic platform for AI infrastructure, anchored by long-term licensing and bespoke programs with hyperscalers and OEMs. That shift increases both upside (higher royalties per AI-enabled unit) and concentration risk (few customers and geopolitical exposure). For deeper scenario analysis and a synthesis of headline partnerships into revenue-impact models, visit https://nullexposure.com/ for research and investment frameworks.

If you want a concise model mapping these customer relationships to potential royalty outcomes, start here: https://nullexposure.com/.

ARM customer relationships