Google Cloud launches a universal Gemini agent for enterprise work

Google Cloud announced the Gemini agent at its Gemini at Work 2026 event, describing it as a “universal agent for work” meant to handle knowledge work, question-answering, content creation and coding from a single prompt box rather than through separate specialised tools. Google said the agent carries an organisation’s business context into every request, plans the steps a task requires, calls on skills and tools as needed, and connects directly to a customer’s own business systems to carry work through to completion rather than stopping at a draft.

The pitch rested on packaging rather than a new underlying model: Google said the agent automatically selects the best model for a given job, includes built-in cost controls, and delivers results inside the documents, inboxes and developer environments staff already use, alongside the security, administration and governance features enterprise buyers require. The announcement itself was brief and pointed to a fuller Google Cloud blog post for detail; no specific customer figures, pricing or availability date were given in the primary announcement.

The launch extends Gemini Enterprise, the workplace AI platform Google introduced a year earlier to compete with Microsoft 365 Copilot, from a platform for building and deploying agents toward a single default agent that aims to absorb most everyday requests itself. It continues a broader pattern in enterprise AI sales through 2026 of vendors selling a managed, governed agent experience rather than bare model access, competing with Microsoft, Salesforce and a growing field of agent-platform companies for the same corporate budgets.