Microsoft launches VMs based on Cobalt 200 chip in preview, Maia 200 already in production
Announced at Microsoft Build 2026
Microsoft is launching its custom AI chips to address the increasing demand for optimized compute infrastructure and reduce reliance on external chip suppliers. This comes after significant R&D investments and market maturation for AI-specific hardware.
This move signifies major cloud providers taking greater control over their hardware supply chain and optimizing performance for AI workloads, impacting chip manufacturers and the economics of cloud computing. It strengthens Microsoft's position in the AI infrastructure race.
Microsoft will offer VMs powered by its custom Cobalt 200 and Maia 200 chips, providing alternatives to NVIDIA and AMD offerings for AI and general-purpose workloads in its Azure cloud. This introduces new competition and optimization layers in the cloud compute market.
- · Microsoft Azure
- · Cloud service users (cost/performance)
- · Hyperscalers with custom silicon initiatives
- · NVIDIA (short-term market share)
- · AMD (short-term market share)
- · Traditional CPU vendors in cloud
Increased performance and potentially lower cost for AI workloads within Azure.
Accelerated trend of hyperscalers developing proprietary chips, further fragmenting the server chip market.
Long-term geopolitical implications as global compute power becomes more distributed across national and corporate providers, tied to specific hardware ecosystems.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at DataCenter Dynamics