By Rajwa Quasim
Google Cloud introduced a new Gemini agent for work on Thursday, expanding the company’s push to make artificial intelligence capable of completing tasks on behalf of employees. The launch comes as companies including OpenAI and Meta compete to develop AI agents that can independently carry out complex tasks.
The company unveiled its Gemini agent, describing it as a “single, universal agent” for workplace tasks. Unlike conventional chatbots that mainly respond to prompts, the Gemini agent can be given an objective and then plan the steps needed to complete it.
Google said its agent can answer questions, perform knowledge work, create content and even write and run code. It can also connect with a company’s business systems and return completed work directly to the applications employees already use, rather than requiring them to move between different AI tools.
READ: Google employees must share data with AI tool to access health benefits: Report (October 9, 2025)
The agent is designed to operate across Google’s workplace ecosystem, including Gmail, Sheets, Slides, Chat and Calendar. It can also connect with third-party services such as Slack and Microsoft 365. Gemini can be accessed on the web, iOS and Android mobile devices, and Windows and Mac desktops.
One of Google’s key selling points is that Gemini can continue working even after an employee leaves their computer. The agent operates in the cloud, allowing it to maintain context and continue tasks for hours or even days.
The system is designed to create temporary sub-agents for complex assignments. Google said these agents can work on different parts of a task in parallel or sequentially and bring the results together after completion.
According to a blog post published by the company, businesses can also create “coworker agents” that function like digital members of a team, with their own identities and email addresses and access to information provided by the organization.
The agent is also capable of choosing the model most appropriate for a particular task. For instance, it can choose Google’s Gemini models or Anthropic’s Claude models. According to the company, this approach will improve performance while helping businesses control costs.
Regarding security and corporate control, Google said the Gemini agent includes administration, authorization, permission controls, secure sandboxing and other governance features designed for enterprise use.
READ: Sundar Pichai says he is ‘feeling lucky’ to work with Nobel laureates at Google (
“In the last year, nearly 500 Google Cloud customers each processed more than one trillion tokens. Today, nearly 80% of all Google Cloud customers are using our AI products, and nearly 90% of the Fortune 100 use Gemini Enterprise,” the blog stated.
Specialized versions for financial and legal work are already in preview, while the company said versions for sectors including government, health care and retail are being planned.
The launch comes as major AI companies race to develop more capable agentic AI systems. These include OpenAI’s “Dots” and Meta’s “Muse,” both of which were released recently.