By combining the analytical depth of Looker with Google’s Agentic Data Cloud, the potential to transform how we model, interact with, and act on our data appears limitless. This week at Google Cloud Next, we are thrilled to share how we are reimagining the business intelligence (BI) stack with Looker for the AI era, starting with a deeper integration with BigQuery.
Our product strategy is focused on delivering trusted data and prescriptive insights for everyone to take actions through Agentic BI.
We’ve rebuilt Looker with Gemini to power a new era of AI agents. With Looker providing the trusted data foundation, your team can innovate rapidly without losing accuracy.
Conversational BI and agents: From conversations to actions
At Next ‘26, we are announcing Looker BI Agents that don’t just provide static answers, but trigger downstream business actions grounded in the Looker semantic layer and your existing enterprise governance framework.
Today, we introduced a number of new agents in Looker:
Dashboard Agents – getting answers to your data directly within a dashboard.
Conversational Agents upgrades (Preview): We are significantly upgrading our core Conversational Analytics agent, which is already GA, with superior reasoning and semantic grounding, helping to eliminate ambiguity. New visibility tools provide admins with end-to-end observability, so that they can monitor performance trends and refine model accuracy at scale. Teams can now also publish Conversational Analytics in Looker directly into Gemini Enterprise to incorporate trusted data into broader workflows.
Dashboard Agents (Preview): Dashboard Agents bring conversational capabilities directly into your BI workflows, providing instant summaries and the ability to ask questions directly within a dashboard.
Embedded Conversational Experiences (GA): You can now embed Conversational Analytics agents directly into your custom applications and internal workflows, allowing users to query data via the UI or API without leaving their specialized tools.
Agentic Workflows in Looker (Preview): Agents can now monitor critical metrics for irregularities and identify hidden correlations and “what’s next” recommendations, so you can address shifts in the business — before they impact the bottom line.
“At YouTube, we’re focused on helping creators succeed and bring their creativity to the world. We’ve been testing Conversational Analytics in Looker to give our partner managers instant, actionable data that lets them quickly guide creators and optimize creator support.” – Thomas Seyller, Senior Director, Technology & Insights, YouTube Business
Reimagining self-service: AI power, UI control
AI powered Google-easy exploration for ease and simplicity.
We are making Looker more intuitive and easy for both business users and analysts.
Additionally, we are introducing new AI assistants in Looker Explores to erase the hard parts of BI and transform complex workflows into natural language interactions that feel as familiar as a Google search. New today are:
The new and modernized Looker interface (Preview): A modernized, Google-easy drag-and-drop interface for AI-driven self-service Explores, featuring tabbed dashboards (GA) and paginated reporting (GA).
Visualization Assistant (GA): Use natural language to create and refine charts. Simply ask to change drafted visuals to a stacked bar chart or color-code by region, and Gemini handles the rest.
Expression Assistant (Preview): Generate custom dimensions, measures, and filters just by describing your desired logic — no more having to remember syntax!
Insight Assistant (Preview): Beyond just showing data, Insight Assistant automatically generates summaries and highlights key trends within your reports, helping you identify the so what in seconds.
Self-service Explores (GA): Blend personal data from CSV and Excel files with your enterprise data, while a new interface makes it easier to build models visually. Behind the scenes, Looker handles the LookML code. Together, they provide the agility of a spreadsheet with the rigorous version control of an enterprise platform, so that even ad-hoc data analysis remains governed.
“Our users (or customers) love the flexibility of Excel but fear the lack of governance. Looker’s Self-service Explores give them the best of both worlds. They can upload ad-hoc CSVs for a quick ‘what-if’ analysis, but because that data is analyzed alongside our LookML dimensions, we know the underlying business logic is still accurate.” – John Pettit, Chief Technology Officer, Promevo
A trusted and open platform
Having a rock-solid, governed semantic layer is a strong way to prevent AI hallucinations and ensure a single source of truth, and we have some exciting announcements in this area:
Open BI and MCP (Preview): Looker supports open-source Model Context Protocol (MCP) via MCP Toolbox, and now we’re offering an all new managed MCP server that’s native to Looker, for ease of customer management. Through this open BI approach you have the flexibility to choose how you deliver AI transformation to your users.
Looker Extension for VS Code (Preview): We are accelerating the developer lifecycle with a new intelligent setup wizard and a specialized LookML AI Agent that translates business intent directly into production-ready code. This new plugin for VS Code and Agentic IDEs lets you vibe-code with LookML, delivering an AI-powered authoring experience for building LookML models simply by describing your desired outcome.
Updated LookML with Gemini support in the Gemini-powered IDEs
Continuous Integration (CI/CD) (GA): Now generally available, this fully integrated advanced CI capability in Looker automates SQL validation and content testing within your development workflow. By identifying potentially breaking changes before they reach production, Looker helps ensure model updates are reliable and accurate.
Knowledge Catalog integration (Preview): With Knowledge Catalog, Looker transforms metadata into a semantic graph, providing the essential context AI agents need to complete tasks autonomously.
The future of data isn’t just about seeing insights — it’s about acting on them. By combining the power of Gemini with Looker’s trustworthy semantic layer, we are making data-driven actions a reality for every user, in every workflow, every day.
Welcome to the era of Agentic BI.
Get started in your journey to transform your business through the agentic era with Looker at cloud.google.com/looker or start your developer experience at https://docs.cloud.google.com/looker/docs/api-sdk.