Looker Launches Managed MCP and Agentic BI Features at Next '26

Full analysis

Google announced a suite of agentic business intelligence (BI) capabilities for Looker at Google Cloud Next 2026, anchored by a 'Managed MCP' (Model Context Protocol) offering that exposes Looker's semantic layer — a centralized, governed definition of business metrics — to external AI platforms. PayPal is cited as a production customer, having scaled conversational analytics to 3,000+ users via Anthropic's Claude Desktop connected to Looker via MCP. Additional features include Dashboard Agents powered by Gemini 3 that let users query data in natural language, a LookML Agent for AI-assisted semantic modeling inside VS Code, expanded support for BigQuery Graph and Snowflake Semantic Views, and a Conversational Analytics API for embedding governed data agents in third-party applications.

The overarching positioning is a shift from reactive dashboards to autonomous agentic workflows: Looker's LookML governance layer is framed as the trust foundation that keeps AI agents grounded in verified metrics rather than hallucinated outputs. Google also announced GA availability of paginated reporting and tabbed dashboards, and highlighted the Gemini-powered Visualization and Insight Assistants for self-service analytics. The company cited a third consecutive year as a Leader in the Gartner Magic Quadrant for Analytics and BI Platforms.