To begin, could you introduce Arm and explain your role within the company?
My name is Will Abbey, and I’m the Chief Commercial Officer at Arm. I’ve been with the company for 21 years and am a technologist at heart. My team helps partners make the right technology choices, build out their technology stacks, and deploy them efficiently. When you think about compute—the brains behind the world’s electronic devices—Arm CPUs sit at the centre of it all. They’re the intelligence that brings these devices to life.
From our founding to today, Arm technology has shipped in over 325 billion devices, with about 33 billion added each year . Whether it’s classic compute or accelerated, AI-driven workloads, Arm remains at the heart of innovation—powering intelligence across virtually every connected device on the planet.
Arm has hinted at designing chiplets—a more holistic approach. What problem are you trying to solve by doing this, and how do you ensure collaboration with partners?
To answer that, it’s important to look at what’s happening in compute. Across the industry, we’re seeing an insatiable demand for computing power, driven primarily by AI workloads. Models, agents, and applications all require more compute, and this exponential growth demands a smarter approach than simply adding more servers or consuming more power. We see customers looking for faster paths to market.
So we’re looking at the full technology stack for our customers—creating solutions that enable high-performance, scalable , power efficient compute.
Some have said there might be a bubble forming in AI and computing infrastructure. How do you view this, and how do you model the next few years of growth?
There’s been a lot of discussion about a potential bubble, but from my perspective, we’re just at the beginning of the AI revolution.
History offers useful parallels: during the internet boom, there were both winners and failures, yet out of that era came giants like Amazon. We see AI following a similar trajectory—some volatility, yes, but long-term transformation that’s real and enduring.
The scale of investment from cloud service providers and leading tech companies demonstrates confidence. People don’t make multi-billion-dollar bets just on hype. As workloads shift from classic to accelerated compute, Arm’s architecture becomes even more relevant. We’re seeing a foundational shift that will define the next generation of computing.
You’ve said Arm fundamentally believes in the AI revolution. Why is that belief so strong?
It’s both professional and personal. Take ChatGPT—six months ago, I used a web browser for every search; now I barely do. AI has reshaped my daily habits, from how I access information to how I interpret my own medical data. I recently uploaded my blood test results into a model to get immediate insights, something unthinkable a few years ago.
Professionally, within Arm, more than 80 percent of employees now use AI tools daily to boost productivity, improve documentation, and streamline verification. We’ve seen measurable efficiency gains, and this is just the start. As AI reshapes industries, power efficiency becomes critical.
Energy is a scarce resource, and Arm’s designs, optimised for performance per watt, ensure that intelligence can run efficiently from the cloud to the edge, a huge advantage as applications demand faster, more responsive compute.
Can you explain how Arm’s v9 and related technologies make AI more efficient across devices and where you’re seeing real-world impact?
Running AI locally depends on power efficiency, compute density, and effective thermal management—all core strengths of Arm’s architecture. These principles, established in our early work with handheld devices, now scale to the cloud and automotive sectors. We’ve always been focused on doing more with less energy, and that philosophy has never been more relevant.
A great example is our collaboration with Meta. From their AI glasses to the data centre, they rely on a common Arm architecture that enables seamless scaling of models across devices. A model trained in the cloud can run efficiently in smaller form on a wearable, without architectural changes. That reduces time to product and accelerates innovation—something essential as AI becomes ubiquitous.
Every AI system, whether it’s an NVIDIA Grace Blackwell or Vera Rubin, has a CPU at its core. Arm is foundational to that, delivering efficient, scalable compute that powers AI workloads across training, inference, and experimentation.
Given recent developments between the U.S. and China, how is Arm adapting its China strategy amid export rules and supply chain tensions?
Arm operates with strict compliance in mind and always adheres to international export regulations. We’ve worked closely with governments worldwide for more than 30 years, and that won’t change. China’s ecosystem mirrors others globally—it has cloud service providers like Alibaba, major automotive innovators like BYD and Xiaomi, and a thriving model-development community.
The demand for performance and innovation in China is no different from that in the U.S. or Europe, and we continue to work closely with partners across all regions while ensuring compliance.
As AI compute shifts from data centres to the edge, how does Arm maintain a competitive advantage over RISC-V and players like NVIDIA?
Arm’s competitive edge lies in efficiency, scalability, and accessibility. Technically, we’re the only company that offers scalable solutions from a single CPU to thousands, optimised for performance, density, and thermals. But it’s also about how partners access our technology. Our business model—through Arm Flexible Access (AFA) and Arm Total Access (ATA)—gives startups and OEMs flexible entry points to explore and deploy technology before full licensing.
We now have over 300 AFA partners delivering real products, and that flexibility, combined with our proven architecture, gives Arm a unique position. The strength of both the technology and the licensing model ensures that innovation can flourish across ecosystems, from early-stage companies to hyperscalers.
Looking ahead, as the world builds out this massive AI infrastructure, what could go wrong? What are the biggest risks the industry faces?
Power is the first and most pressing challenge. We’re already seeing scenarios, such as in parts of Scotland, where municipalities must choose between allocating energy to schools or data centres. If the industry doesn’t innovate around power efficiency, we’ll hit hard limits. That’s why we’re pushing for intelligent use of power and performance—building smarter, not just bigger.
Security and AI ethics also stand out. As AI systems expand, bad actors will find new vulnerabilities, making embedded security essential. Arm’s v9 architecture includes features like memory tagging and other built-in protections to address this. Lastly, we must be mindful of AI ethics—how we train models and use data responsibly. These three pillars—power, security, and ethics—are critical for sustainable progress.
When you think about the next generation of engineers and leaders, what skills will matter most?
We look for curiosity above all else—engineers who question assumptions and explore deeply. Technical excellence is vital, but so is comfort with data science and an understanding of machine learning. Engineers must be able to navigate these tools creatively.
Soft skills are equally important. The ability to communicate complex ideas clearly within diverse teams is what turns technical skill into impact. In our experience, curiosity, technical depth, and communication form the foundation of great engineering leadership.
Finally, you’ve had an international career across Ghana, the UK, and now Silicon Valley. How has that global experience shaped your perspective?
It’s shaped everything about how I lead. Having lived and worked in Ghana, the UK, and the U.S., I’ve developed an appreciation for cultural nuance and diverse leadership styles. Curiosity has guided me throughout—from a young boy fascinated by technology to a leader who values mentoring and authenticity.
I’ve learned that bringing your differences into the workplace isn’t a weakness—it’s a strength. Whether I’m in northern England, Asia, or North America, I focus on people first. Authenticity and diversity of thought create stronger teams and better innovation. That’s something I try to embody and encourage in every part of Arm.
