Microsoft launches Surface Ultra laptop powered by Nvidia RTX Spark

Microsoft officially launched the Surface Ultra, a high-end Windows laptop powered by Nvidia's RTX Spark chip, in a move to establish a dedicated category for devices capable of running demanding AI workloads locally. The new hardware starts at $2,599 and is scheduled to become available on October 16.

The launch event took place in San Francisco, attended by Microsoft CEO Satya Nadella, Nvidia CEO Jensen Huang, and former White House AI official Sriram Krishnan. Huang described the release as a reinvention of the computer, stating that the company has redefined the device as it is currently known. This announcement follows a preview in June where Microsoft had withheld specific details regarding pricing and exact availability.

The Surface Ultra is the first of several Nvidia-powered devices entering the market. Microsoft confirmed that similar hardware will also be produced by Asus, Dell, HP, and Lenovo, with all these models set to arrive on the same October 16 start date. The strategy aims to serve users who prefer to process complex AI tasks on their own machines rather than relying exclusively on cloud-based services.

Alongside the hardware, Microsoft introduced a version of its Copilot assistant that functions using both local and cloud-based models. The company also announced that Meta will bring its Muse personal agent to the Windows platform. In outlining its security framework, Microsoft emphasized measures designed to separate actions taken by AI agents from those performed by humans, ensuring that agent behavior remains within preset limits.

The push into AI-powered Windows machines marks a second phase for Microsoft after the initial rollout of Copilot+ PCs encountered difficulties. The first generation faced a rocky start, primarily due to delays in launching Recall, a signature feature of the platform. Microsoft executive VP Pavan Davuluri explained that hybrid intelligence is necessary because user demand for capabilities exceeds current budgets for frontier cloud models. Davuluri noted that this approach allows for the use of local models when effective, complementing cloud-based resources.

The company's positioning suggests that developer appetite for AI capabilities outpaces what corporate budgets can support through cloud access alone. By combining on-device processing with cloud resources, Microsoft seeks to address this gap while maintaining strict security boundaries for autonomous software actions.