Arm Holdings

Arm Holdings’ stock surged 20% after announcing its new AGI CPU, designed for AI inference in data centers. The company projects the chip will generate $15 billion in revenue by 2031, a significant shift from its traditional IP licensing model. This move aims to capture a larger share of the growing AI market, with early adopters including Meta and OpenAI.

Arm Holdings is pivoting from IP licensing to chip manufacturing with its new AGI CPU, targeting the booming agentic AI market. CEO Rene Haas projects $25 billion in annual revenue by 2031, a significant leap from current figures, with Meta as the first customer. This strategic shift aims to capture greater value in the AI hardware sector, offering customers more choice and expanding Arm’s addressable market.

Arm Holdings is shifting its AI focus from cloud to edge computing, anticipating a significant market transformation. Vince Jesaitis, Arm’s Head of Global Government Affairs, highlights the advantages of localized AI processing: enhanced power efficiency, reduced latency, and improved data privacy. Arm’s low-power chip designs are ideal for edge AI, supporting enterprise digital transformation and meeting ESG goals. The company is also engaging with governments on workforce development and regulatory approaches, positioning itself to power both large providers and the growing edge AI demand.

Arm CEO Rene Haas advocates for distributing AI workloads from cloud-based infrastructure to local devices to reduce energy consumption and improve sustainability. He highlights a shift towards hybrid computing, with AI training in the cloud and inference occurring on devices like smartphones and AR glasses. Arm’s expanded partnership with Meta aims to optimize AI efficiency across the entire compute stack, exemplified by localized speech recognition in Meta’s Ray-Ban Wayfarer glasses. This localized processing enhances responsiveness and reduces reliance on cloud servers.

Arm Holdings plans to launch its dedicated Neural Processing Unit (NPU) for GPUs in 2026, with devices expected by late next year. Simultaneously, Arm unveils the open-source Neural Graphics Development Kit to streamline integration. A key application is Neural Super Sampling (NSS), which upscales resolution with low latency and near-native quality, reducing GPU workload by up to 50%. This innovation promises improved power consumption, higher frame rates, and enhanced visuals for gaming and other applications.