Arm

“Giving people time back” is one of the opportunities at the heart of physical AI. In a conversation with Austin Lyons from Creative Strategies, Inc., Drew Henry explores the broader potential of bringing AI into the physical world, from driving productivity across transportation, manufacturing and logistics to tackling the complex systems challenges behind robotics and autonomous vehicles. The discussion also looks at the growing demands on compute, how workloads will span edge to cloud, and how we're enabling the flexibility and software compatibility needed to build across a diverse range of physical AI systems. 🎧Watch the full conversation here: https://okt.to/YFC0oG

Moving AI agents from pilots to production changes the infrastructure equation. Agents need secure environments to execute code; those environments need to spin up quickly, and workloads can range from short bursts of code to long-running, compute-intensive jobs. In our conversation with Google Cloud and The New Stack, we explore how Arm-based Google Axion processors and GKE Agent Sandbox help developers run and scale these emerging agentic workloads. ⚡️ https://okt.to/1nqCk0

The next frontier for AI is the physical world. AI that can sense, reason and act safely in the real world—robots, vehicles, drones and other autonomous machines. But there’s an important difference between Physical AI and the AI we’ve spent the last several years building in data centers. An AI Factory operates in a highly controlled environment with effectively unlimited power, cooling and weight. Physical AI has to bring enormous amounts of intelligence into machines constrained by power, weight, heat and cost—while operating in an unpredictable world, often with real-time and safety-critical requirements. In many ways, I think that makes the computing challenge for Physical AI even more complicated than building an AI Factory. And that’s exactly why I believe Arm is in the middle of this transition. I recently sat down with Austin Lyons from Creative Strategies, Inc. to discuss where Physical AI is headed, what it demands from computing, and the role Arm can play in enabling it. https://lnkd.in/edUMY-2y

OCP APAC is where the AI infrastructure ecosystem comes together to tackle what’s next. 🚀 Through workshops, breakout sessions, and conversations with the community, we’ve been digging into what it takes to build more open, scalable infrastructure for the AI era. Arm SystemReady put standards-based interoperability in focus, while sessions on the Arm AGI CPU explored the evolving role of the CPU as AI infrastructure scales. Because scaling AI infrastructure goes beyond compute alone—it requires innovation across the technology stack, with open standards and interoperability helping hardware and software ecosystems work together as infrastructure evolves. Together, we’re advancing the technologies, standards, and system-level innovation needed to move AI infrastructure forward.

📱 The next generation of Google Pixel has arrived. The new #Pixel11 series continues Google’s innovation on Arm, bringing together advanced compute and AI to power the next generation of mobile experiences, designed for Gemini Intelligence. We’re proud to see Arm technology at the foundation of Google’s latest Pixel devices, helping deliver the performance and efficiency needed as AI becomes an increasingly integral part of the smartphone experience. Congratulations to the Google team on another exciting #MadeByGoogle launch! 👏https://okt.to/ZaJePQ

AI is changing what modern data centers demand from the CPU, and at OCP APAC in Taiwan this week, Michael Wong, President of Arm Taiwan & SE Asia, took the stage to explore what that evolution means for AI infrastructure. His keynote looked across the technology stack, from Arm-based compute and open standards to system-level co-design spanning: ➡️ Silicon ➡️ Chiplets ➡️ Packaging ➡️ Software ➡️ Systems Building AI infrastructure at scale requires an ecosystem approach. Bringing the industry together around open collaboration helps accelerate innovation and create scalable, interoperable infrastructure built for the demands of AI.

The industry has spoken: AI infrastructure is being built on Arm. Arm-based rack-scale GPU servers have overtaken x86 in accelerated server spending, according to IDC. This milestone confirms Arm’s position at the foundation of the world’s AI infrastructure. Our platform delivers the performance, efficiency and scale needed to power AI—from the largest data centers to the edge. Arm is the compute platform for AI. We’re just getting started. https://okt.to/OiRdzL

Where will your AI make the biggest impact? Whether you're building intelligent robots, scaling AI inference in the cloud, or creating on-device AI experiences, the Arm AI Optimization Challenge has a track for you. Explore the three challenge tracks, Physical AI, Cloud AI, and Mobile AI and discover the kinds of projects you can build, optimize, and submit. We're looking for projects that demonstrate measurable AI optimization on Arm-powered platforms. Compete for a share of the $8,000 cash prize pool, showcase your work to the global developer community, and put your optimization skills to the test. ⏰ Submissions close 14 August. https://okt.to/rnjzMw

Arm has long believed that open source gives developers more choice in how they build AI. We’re excited to see AI at Meta expand that ecosystem with its new open-weight model Muse Glimmer, bringing powerful local, agentic AI experiences closer to developers and devices. We look forward to enabling these experiences across Arm-powered platforms. 🎉

How do you take something that was never designed to operate as a manufacturing company and turn it into one that can deliver, at scale, with quality, on time? That’s the challenge Eric Hayes, EVP of Operations here at Arm, took on while helping build the operational capabilities needed to expand from IP and Arm Compute Subsystems, to now include silicon production. Now he’s pulling back the curtain on what it took to build the operations, supply chain, and capabilities behind this shift, and make this next era of the Arm compute platform possible. 💡 https://okt.to/hEkbpc