Google Cloud unveils persistent Gemini Agents for long-running tasks, and they get their own Gmail, Calendar, and Drive storage
Google is unveiling a new universal AI agent for the workplace that can research information, create documents, write code, coordinate other AI agents and complete tasks lasting hours or days — all while maintaining the same memory, business context and identity across different applications and…
Google Cloud has introduced a new universal AI agent for the workplace, designed to handle long-running tasks across multiple applications and devices while maintaining consistent memory and context.
What happened
Google Cloud unveiled a new universal AI agent for enterprise use, capable of handling long-running tasks across various applications and devices. The agent can function as a persistent digital coworker with its own identity and resources, and it supports multiple AI models. The announcement highlights Google's efforts to unify its enterprise AI offerings and compete with other companies like Microsoft and OpenAI.
Why it matters
This development is significant for AI developers and enterprise IT managers as it introduces a new approach to managing AI tasks in the workplace. The ability to handle long-running tasks and maintain context can improve efficiency and reduce the need for manual intervention. The integration with multiple AI models and enterprise systems also offers flexibility and scalability for businesses.
- Google's new AI agent offers a unified approach to enterprise AI tasks with cross-platform capabilities.
- The agent can function as a persistent digital coworker, improving task management and efficiency.
- Support for multiple AI models provides flexibility for enterprises.
- Security and access controls are emphasized to protect sensitive data.
- The competition in the enterprise AI space is intensifying, with multiple companies offering similar solutions.
The bigger picture
The introduction of Google's universal AI agent could streamline enterprise workflows by enabling persistent, context-aware AI coworkers. This may lead to increased efficiency in task management and reduced operational overhead. However, the success of this initiative depends on the reliability of the system, the availability of features, and the ability to manage costs effectively.
Google Cloud's new universal AI agent represents a significant step towards integrating AI into enterprise workflows, offering persistent, context-aware capabilities across multiple applications and models. While the announcement highlights potential benefits, questions remain about feature availability, reliability, and cost management.
Technical details
- Universal AI agent with cross-platform memory and context retention.
- Support for multiple AI models (Gemini, Claude, and others).
- Integration with enterprise systems like Salesforce, Jira, and Microsoft Teams.
- Security features including cryptographically attested identities and audit logs.
- Agent Gateway for enforcing organization-wide policies.
- Smart Routing for selecting models based on performance and cost.
FAQ
What is Google's new universal AI agent?
Google's new universal AI agent is a workplace tool designed to handle long-running tasks across multiple applications and devices, maintaining consistent memory and context.
How does the AI agent function as a digital coworker?
The AI agent can operate as a persistent digital coworker with its own identity, email, calendar, and storage, allowing it to manage tasks and interact with employees as a team member.
What AI models does the agent support?
The agent supports Google's Gemini models and can also use Anthropic's Claude models, with plans for additional proprietary and open-weight AI models.
What are the security features of the AI agent?
The agent includes security features such as cryptographically attested identities, audit logs, and access controls to ensure it only accesses shared information and follows organizational policies.
How does the agent compare to competitors like Microsoft and OpenAI?
Google's agent is similar to Microsoft's Copilot and OpenAI's Dots agents, but it emphasizes cross-platform integration, security, and support for multiple AI models.
This summary and analysis were generated with AI from the original article and may contain errors. Read the original for the full story.
Read the full story
Originally published by VentureBeat AI on Oct 8, 2026.
Comments
Sign in to join the discussion.