News

Google Packages Enterprise AI Agents into New Gemini Platform

• By John K. Waters
• 04/30/2026

Google Cloud has introduced a new platform for building and managing enterprise AI agents, as the company seeks to turn its Gemini models and Vertex AI tooling into a broader system for automating business workflows.

The new product, called Gemini Enterprise Agent Platform, was announced at Google Cloud Next ’26 and is described by the company as an evolution of Vertex AI. Google said the platform combines model selection, model building, and agent-building capabilities with newer tools for agent integration, DevOps, orchestration, governance, optimization, and security. Search on "Vertex AI" and you get "Gemini Enterprise Agent Platform (formerly Vertex AI)," so it's a significant rebranding.

The launch reflects a shift in the enterprise AI market from chat-based assistants to agent systems that can perform multistep tasks across corporate applications, data sources, and internal processes. Google is positioning Gemini Enterprise as an end-to-end system for what it calls the “agentic era,” in which companies delegate business outcomes to AI agents rather than use them only for isolated tasks.

Google said the platform is designed to help companies build, scale, govern, and optimize agents. In practical terms, that means providing tools to connect agents to enterprise systems, deploy them through development workflows, monitor their behavior, apply security controls, and improve performance over time.

The company is also expanding the ecosystem around Gemini Enterprise. Google said partner-built agents from its Agent Marketplace will be available within an Agent Gallery in the Gemini Enterprise app, giving customers access to specialized agents from companies such as Adobe and Atlassian.

Google also announced a $750 million innovation fund for partners developing and deploying AI agents. The fund aims to encourage partners to build agents for business processes, functions, and industries, underscoring Google’s effort to make Gemini Enterprise a platform for third-party development as well as for its own AI services.

The announcement comes as large cloud and software companies race to define the market for enterprise AI agents. Microsoft, OpenAI, Anthropic, Salesforce, ServiceNow, and other vendors are all trying to persuade companies that their platforms can safely automate work across sales, customer service, software development, finance, human resources, and operations.

Google used Cloud Next to argue that enterprise adoption is already moving beyond experiments. The company said nearly 75% of Google Cloud customers are using its AI products, and that its models now process more than 16 billion tokens per minute via direct customer API calls, up from 10 billion in the previous quarter.

The more significant claim behind the announcement is that enterprise agents will require infrastructure, not just models. Companies that deploy agents at scale will need identity controls, audit trails, policy enforcement, integrations with existing software, monitoring tools, and mechanisms for testing and updating agents after deployment.

That is where Google is trying to differentiate the Gemini Enterprise Agent Platform. Rather than presenting it as a single assistant, Google is packaging it as a control layer for many agents operating across an organization.

The strategy also gives Google a way to extend Vertex AI into a broader enterprise product category. Vertex AI has been Google Cloud’s primary platform for building and deploying machine-learning and generative AI applications. By framing the Gemini Enterprise Agent Platform as its evolution, Google is signaling that agent development is becoming a core part of its cloud AI business.

For customers, the pitch is straightforward: build agents using Google’s models and tools, connect them to business systems, manage them under enterprise controls, and add partner-built agents where useful.

The risks are equally clear. Many enterprises remain cautious about giving AI systems access to sensitive data or authority to act inside business workflows. Reliability, accountability, compliance, cost, and security remain barriers to wider deployment, especially for agents that do more than summarize information or draft text.

Google’s announcement does not resolve those questions. It does, however, show where the enterprise AI market is heading. The competition is no longer only about which company has the strongest model. It is increasingly about which vendor can provide the safest and most useful environment for running fleets of AI agents inside large organizations.