Google Cloud has introduced Gemini agent, a universal AI tool designed to function as a digital employee. These agents operate within the cloud, allowing them to continue tasks even when a user is offline. Each agent is assigned a corporate email address, calendar access, Drive storage, and a position in the company directory.
The system integrates chat, task execution, and coding into a single interface. Users define a goal, and the agent plans and executes the work. For each task, the platform selects an appropriate model, currently supporting Gemini and Anthropic’s Claude, based on quality and cost requirements.
To ensure accountability, each agent receives a cryptographically verified identifier. All actions are logged in an audit trail attributed to the agent rather than a human user. Agents operate in isolated environments, and companies can set spending limits to automatically pause operations if costs exceed defined thresholds.
Industry-specific versions for the legal and financial sectors are currently available in preview mode. Google has not yet disclosed pricing or a timeline for a general release. Previous internal research by the company highlighted potential social risks, noting that agents previously made errors such as sharing private drafts or sending inappropriate messages to executives.
