Skip to content
Status: Generally available Announced in BRK423

AI Gateway for unified AI management and policy

Ignite 2026

AI Gateway, for unified AI management and policy, was announced as available today in session BRK423.

Sources

BRK425 Breakout Wed 18 Nov 13:45 PST

Secure and manage AI workloads at enterprise scale with AI Gateway

Mike Budzynski, Facundo Santiago, Raghu Balam

From the recording

Microsoft announced Azure AI Gateway as generally available, stating this took effect the day before the session. It is available in 16 Azure regions, with more being added, and supports automated scaling and private connectivity. Auto model selection, which uses the Microsoft Foundry router, is in preview. The gateway mediates agent traffic to models and MCP servers, with Entra authentication, layered policies and OpenTelemetry observability. Demos covered alias-based failover and blocking a leaked secret. Caveat: pricing, exact GA date and whether GA covers every demonstrated feature were not stated; captions are auto-generated.

THR902 Theater Thu 19 Nov 17:30 PST

Take control of runaway AI costs with AI Gateway

Mike Budzynski

From the recording

Mike Budzinski, a Microsoft Principal Product Manager, presented Azure AI Gateway, which he said became generally available two days before the session. It sits between AI clients and models in Foundry, AWS Bedrock, OpenAI, Anthropic or Google Cloud. The demo covered OpenTelemetry cost monitoring, Azure RBAC model access, a cost limit policy, aliases with priorities, weights and session affinity, and auto model selection. He said it is free to start with. Caveat: cost figures are estimates from observed tokens and public pricing, and the demo used fictional models.

More in this area

← All announcements