Azure AI Bot Service

About Azure AI Bot Service

Azure AI Bot Service is Microsoft's Azure service for registering, connecting and operating conversational bots across supported channels. It is still a current Azure offering, but buyers should separate the service itself from the older Bot Framework SDK tooling. Microsoft has archived the Bot Framework SDK and Emulator and now recommends Microsoft 365 Agents SDK for modern agent development, while existing Azure Bot resources can continue to provide channel connectivity and bot registration. This makes Azure AI Bot Service most relevant for teams that need Azure-hosted bot connectivity, channel integration and continuity for existing bot estates while they plan newer agent architecture carefully.

What does Azure AI Bot Service do today?

Azure AI Bot Service provides the Azure-side service used to connect bots with communication channels and manage the bot registration that routes activities between a bot application and channels. Microsoft's current product page positions the service for enterprise conversational experiences and highlights connections to websites, mobile apps, Microsoft Teams and other supported channels. Standard channels and premium channels use different pricing rules, and the bot's own application hosting is billed separately when it runs on Azure App Service or another Azure compute service.

The important architectural point is that Azure AI Bot Service is not the same thing as the application code that implements the conversation. A team can operate an Azure Bot registration while hosting the bot logic in an application it controls. That separation matters when estimating cost, planning migration and deciding whether newer Microsoft 365 Agents SDK or Copilot Studio tooling should replace older Bot Framework development patterns.

How is Azure AI Bot Service priced?

Pricing was checked on August 29, 2026. Microsoft's current pricing page lists Standard channels with unlimited messages and a Free tier for the Bot Service channel layer. Premium channels are metered by messages, with the Free tier including 10,000 premium-channel messages per month and paid S1 usage billed per 1,000 messages. Microsoft does not expose one universal final price because rates vary by agreement, currency and region.

The channel charge is only part of total cost. Microsoft states that an Azure Bot can provision an Azure Web App to host the bot, and that Application Insights may also be used for telemetry. Those resources have their own Azure pricing. Language, speech, AI model or data services used by the bot can add further charges, so buyers should model the entire solution rather than treating Bot Service pricing as the total bot cost.

What channels and integration patterns should buyers consider?

Microsoft's product material describes multichannel delivery across websites, mobile applications, Microsoft Teams and other supported destinations. Direct Line and Web Chat are important when an organization wants to embed a bot into its own application or website, while Teams is a common enterprise channel. Premium-channel billing applies to certain channel scenarios, so channel choice affects both architecture and cost.

Channel support does not remove the need to test authentication, identity, message formats and channel-specific behavior. A bot that works in Web Chat can still need separate validation in Teams or another destination. Organizations should test each production channel independently and include failure handling, telemetry and operational ownership in the design.

What changed with Bot Framework SDK and what should new projects use?

This is the most important current lifecycle consideration. Microsoft says the Bot Framework SDK and Bot Framework Emulator have been archived on GitHub and are no longer updated or maintained. Support tickets for the Bot Framework SDK stopped being serviced after December 31, 2025. Microsoft's migration guidance now recommends Microsoft 365 Agents SDK for modern agent development.

That does not mean every existing Azure Bot resource must be removed. Microsoft's migration guidance explains that existing Azure resources can remain in place during migration, including the current Azure Bot registration, app ID and credentials. Teams with an established Bot Framework implementation should therefore treat migration as an application modernization project rather than deleting Azure Bot resources or recreating channel registrations without a plan.

How does Azure AI Bot Service relate to Microsoft 365 Agents SDK and Copilot Studio?

Azure AI Bot Service is primarily the Azure service and channel-registration layer. Microsoft 365 Agents SDK is the newer development framework for building agents with modern conversation management and orchestration. Copilot Studio is Microsoft's hosted low-code platform for building and managing conversational agents with less custom code.

For a new project, buyers should first decide whether they need a low-code managed experience, a pro-code agent application or only channel connectivity for an existing workload. Copilot Studio can reduce custom development for common business scenarios. Microsoft 365 Agents SDK gives developers more control over application code and integration choices. Azure AI Bot Service remains relevant where Azure Bot registration and channel connectivity are part of the deployment.

What are the practical limitations and migration risks?

The main limitation is architectural age around older Bot Framework patterns. Existing Bot Framework bots can continue to run, but the SDK itself is archived and no longer maintained. Features tied closely to older Composer artifacts, Adaptive Dialogs and other legacy packages may require redesign when moving to Microsoft 365 Agents SDK. Microsoft's migration guidance explicitly lists capabilities that do not map directly and may need alternatives.

Another limitation is cost fragmentation. Bot Service channel pricing, application hosting, telemetry, AI services and data stores can all be billed separately. Channel behavior also differs, so testing one channel does not prove every channel will behave identically. Teams should inventory dependencies before migration and avoid assuming that swapping SDK packages is the only change required.

Who should choose Azure AI Bot Service?

Azure AI Bot Service is a practical fit for organizations that already operate Azure Bot registrations, need established channel connectivity, or want to connect a custom conversational application to channels such as Teams, Web Chat or Direct Line. It can also make sense during a staged modernization where channel registrations remain stable while application code moves from Bot Framework SDK to Microsoft 365 Agents SDK.

It is especially relevant when the organization already uses Azure for hosting, identity, monitoring or AI services and wants bot operations to sit inside the same cloud environment. Buyers should still evaluate whether the bot itself should be built with Microsoft 365 Agents SDK, Copilot Studio or another framework.

Who should choose something else?

Choose Microsoft Copilot Studio instead when the priority is a managed low-code agent platform for business users and developers rather than maintaining a custom bot application. Choose Microsoft 365 Agents SDK when building a new pro-code agent and you want Microsoft's current development path instead of starting on the archived Bot Framework SDK.

A team that only needs a website chat widget with simple support automation may find a specialized SaaS help-desk or customer-service platform easier to operate. Organizations that do not need Bot Service channel routing should also avoid adding it merely because their workloads run on Azure. The right choice depends on whether channel connectivity, custom application control and Azure integration justify the additional moving parts.