Who can do it: Company admins
Which plans: Enterprise with AI Workflows add-on
Which platforms: Browser, Desktop
AI provider availability allows Company Admins to control which AI providers and models are available and how they are used across Miro AI features, including content generation, AI workflows, and Sidekicks. This includes AI-powered capabilities such as Docs, Images, Tables, and AI workflows.
This capability helps organizations balance AI adoption with governance, compliance, and risk management by enabling admins to restrict or allow specific AI providers based on internal policies. It acts as a centralized control layer that governs how AI providers and models are used across different AI capabilities in Miro.
AI provider and model support is defined at the capability level. Each capability may support different AI providers and models, or may not support provider availability controls. Support is not uniform across all capabilities, and differences are expected as features evolve.
How AI provider availability works
AI provider availability controls which AI providers are available across supported capabilities. Within those constraints, users can select from the allowed AI providers and models when using supported AI features.
Depending on the capability, AI provider and model availability may be:
- User-selectable: users can choose from available AI providers and models (Select Your Own Model)
- Admin-controlled: admins define the allowed AI provider or model
- Not supported: the capability uses a predefined AI provider and model
Each supported AI capability has a default AI provider and model defined at the capability level, meaning different capabilities may use different defaults. When multiple AI providers are enabled, this default is pre-selected, and users can choose from the available providers and models within that capability. If only one AI provider is enabled, it is used automatically and users do not see provider or model selection options. For capabilities that do not support provider availability controls, a predefined AI provider and model is used.
In supported capabilities, users can select AI providers and models directly within the prompt or configuration interface. Some AI-powered features, such as default format generation, may have different provider and model support compared to structured capabilities like flows or content generation blocks.
Key use cases
Control risk and compliance
Enable only approved AI providers and restrict providers that do not meet internal security, compliance, or governance requirements.
Standardize AI usage across the organization
Ensure consistent AI behavior and outputs by limiting which AI providers and models are available to users.
Test AI providers before rollout
Evaluate new AI providers in a controlled environment before enabling them for all users.
Limitations
- Not all AI capabilities support AI provider availability
- Support varies by capability and feature maturity
- Some capabilities may remain fixed to specific AI providers and models
- Full compliance coverage may not be available during early stages