What’s new in Unity AI Gateway: service policies, guardrails, observability, and cost controls for AI agents and MCPs

Unity AI Gateway introduces new governance capabilities for AI agents and MCPs, including LLM-based guardrails, cost controls, observability, and service policies.

• •LLM guardrails enforce customizable safety and compliance policies on model inputs and outputs without breaking agent workflows
• •Cost controls enable per-user alerts and hard budget limits to prevent runaway costs across multiple models and providers
• •Payload logging captures complete request and response data for MCP interactions, stored as queryable system tables in Unity Catalog
• •Service policies allow admins to define which tools agents can invoke based on identity, user context, and request parameters
• •All capabilities integrate with Unity Catalog for centralized, auditable governance across AI agents, tools, and models