Google Cloud introduced the Gemini Enterprise Agent Platform on April 22, 2026, during the Google Cloud Next conference. This evolution of the Vertex AI platform enables businesses to build, manage, and scale autonomous AI agents.
The platform allows agents to operate independently for up to 7 days. A feature known as Memory Bank enables these agents to maintain context throughout long-running workflows to ensure task continuity.
Agents are integrated with Google Workspace, providing them with individual email, calendar, and Drive storage. This allows them to function as distinct participants in business processes without impersonating human employees.
The system includes the Agent Development Kit for engineers and Agent Studio for low-code development. Users can access over 200 models, including Gemini 3.1 and third-party options like Anthropic's Claude.
Security is managed through cryptographic agent identities, which serve as tamper-resistant digital badges. Google Cloud aims to compete with Microsoft Azure and AWS by leveraging its Workspace ecosystem.
