Google Aims at Agentic Enterprise with Slew of New Products

Author: Beth Schultz, VP of Research and Principal Analyst

What's the news?

On April 22, 2026, Google Cloud introduced a variety of new capabilities across the AI stack, comprising AI infrastructure, research and frontier models, data management and context, secure enterprise foundation, and agents and applications. The announcements came at Google Cloud Next 2026, the company’s annual event for IT, developer, and business professionals. As CEO Thomas Kurian shared during an analyst briefing on the announcements, the big focus is on tools that enable companies to build, manage, deploy, and operate agents securely, with all controls required.

Among product highlights is the introduction of the Gemini Enterprise Agent Platform, which brings together model selection, model building, and tuning services of Vertex AI along with new features for agent integration, security, DevOps, orchestration, and more. It also includes the Gemini Enterprise app. Some highlights include:

• For building AI agents, Google has introduced the Agent Developer Kit, a graph-based framework for defining how agents work together; Agent Studio, a low-code interface for building, testing, and publishing agents using natural language; Agent Registry, for indexing internal agents and tools, making them discoverable for governance; ecosystem integrations via Agent Marketplace for ease of deploying partner agents; and support for Model Context Protocol, allowing connectivity of Google’s Agent Platform to any MCP server.
• For scaling and orchestrating agents, Google has introduced Agent-to-Agent Orchestration. This enables agents to delegate tasks to one another seamlessly, including support for complex, generative, and deterministic orchestration patterns.
• For governance, Google now offers Agent Identity, which designates agents with a unique cryptographic ID and defined authorization policies that are traceable and auditable. Additionally, Agent Gateway provides centralized, real-time policy enforcement and Agent Anomaly Detection proactively flags suspicious behavior, while the Agent Security dashboard provides a full view of AI risks and compliance using the Security Command Center.
• For optimization, Google introduced Agent Observability, which provides a dashboard with automated logging and agent auditing; Agent Simulation, for stress-testing agents at scale; and Agent Evaluation, for continuous scoring of agent performance against live production traffic.

For workplace collaboration, Google introduced a slew of personal assistance capabilities across the suite, built on top of a workplace intelligence foundation—a semantic layer that connects Workspace apps and Gemini data and agents in a unified context. Among them are:

• For Chat, Ask Gemini in Chat provides a unified command line for getting work done, including a daily briefing for surfacing tasks, unread threads, and urgent action items; the ability to generate documents and slides, as well as other complex tasks, schedule meetings, and find files based on description. Additionally, third-party connectors allow Gemini to tap into external tools for task completion
• For Gmail, AI Inbox presents streamlined view of important Inbox items, and AI Overviews in Gmail search, via Gemini, pulls together information from various email threads to summarize key points.
• For Sheets, Gemini in Sheets enables users to build or edit spreadsheets using natural language descriptions, synthesizing data across files, emails, chats, and the web for data visualization
• For Docs, Gemini in Docs now provides the ability to add infographics to documents using business data, edit multiple images simultaneously for visual consistency, and triage and respond to comments in documents, as well as edit the document based on feedback in comments
• For Slides, Gemini enables the ability to create full, editable slide decks that adhere to company templates and visual styles
• For Drive, Google Drive Projects provides an intelligent space for organizing team files and emails in order to facilitate workflow management, content generation, and search with project context
• For Gemini Enterprise, a Workspace agent executes complex, multi-step tasks across the suite without having to leave Gemini Enterprise