Cognizant is turning its Gemini strategy into a major enterprise AI deployment effort. In partnership with Google Cloud, the company has formed the Gemini Frontier Capability Group, bringing more than 18,000 Cognizant associates into a push to build production-grade agentic AI solutions for clients.
The move arrives as companies race to deploy AI assistants and autonomous agents across departments, yet governance infrastructure has developed more slowly than the technology itself. Cognizant’s new group aims to close the distance between promising demonstrations and AI systems that can operate inside complex business environments.
A Dedicated Team for Agentic AI
The Gemini Frontier Capability Group extends work that Cognizant began with a dedicated Gemini practice launched in April 2026. Its mission is direct: accelerate enterprise-grade agentic AI deployment by combining Cognizant’s engineering and integration expertise with Google Cloud’s AI platform.
“Establishing our dedicated Gemini Frontier Capability Group lets our best frontier engineering talent focus entirely on getting frontier AI into production for clients,” said Ravi Kumar S, CEO of Cognizant.
Initial engineering efforts will focus on four areas that sit at the center of enterprise transformation:
- Legacy modernisation: updating older systems so they can support new AI-driven workflows.
- Developer experience: helping development teams build and use AI capabilities across their work.
- Autonomous business process agentification: turning business processes into workflows that AI agents can manage.
- Frontier cyber defence: applying advanced AI to protect systems and respond to security demands.
These priorities point to a practical goal. The group is not focused only on creating intelligent tools; it is working to connect AI with the systems, processes, and security controls that businesses already depend on.
From Workshops to Large-Scale Deployment
Cognizant has also built a large internal base of Gemini experience. Over 55,000 associates completed the Build with Gemini workshop, giving them direct experience with the platform and tangible skills for enterprise AI work.
“Our Build with Gemini workshop gave over 55,000 associates that experience directly, earning them tangible skills on one of the world’s most advanced AI platforms, and this new group is where that experience goes to work every day,” Ravi Kumar S said.
Cognizant plans to deploy the Gemini platform to 100,000 associates in 2026, with plans to extend access to 200,000 associates. The company also established a dedicated EMEA AI Unit in July 2026, adding another organizational structure for expanding AI work across the business.
Google Cloud CEO Thomas Kurian framed the partnership around the demands of enterprise deployment: “Deploying AI agents at scale requires trusted partners with deep architectural and integration expertise.”
The collaboration already includes several business-focused solutions. Cognizant’s Catalog Support Agent targets retail, while its Healthcare Agentic Marketplace focuses on healthcare. For manufacturing, the companies are developing multi-agent Order-to-Cash workflows that connect activities across a core business process.
Why Governance Must Catch Up
Scaling AI agents creates a new challenge: businesses must understand what these systems do, where they operate, and whether they follow the rules set by the organization. Many enterprises now run AI assistants and autonomous agents across multiple departments, creating activity that can stretch across applications, teams, and approval chains.
Platforms like WalkMe can help governance teams understand workflows and AI activity. That visibility matters because an AI system pulling information from several internal applications must respect access restrictions. Poorly defined approval processes also become harder to trace once automated, especially when several agents handle different steps in the same workflow.
The pressure does not stop at access controls. AI can surface weaknesses in organizational decision-making, creating new demands for the people responsible for oversight. Governance teams need to track human overrides, failed transactions, unauthorized access attempts, model errors, and workflow exceptions to understand whether an AI deployment is working as intended.
Organizations must also account for the resources required to run AI at scale. Those demands affect deployment cost and environmental footprint, making operational planning part of responsible AI management rather than a separate concern.
The Next Test Is Reliable Enterprise Scale
The gap between proof of concept and production deployment remains a pressure point for enterprise AI. A successful demonstration can show what an agent might do, but production requires an inventory of models and agents, clear ownership assignments, defined permissions, and review procedures that remain visible as workflows change.
Cognizant’s Gemini group places that challenge at the heart of a large deployment program. Its engineering focus, internal training effort, and business solutions show how the company plans to move from platform access toward operational use.
Organizations that combine adoption visibility, identity controls, workflow monitoring, and governance processes may be better positioned to scale AI reliably. As Cognizant and Google Cloud push agentic AI into legacy systems, developer tools, business operations, and cyber defence, the defining question will be clear: can these agents deliver useful work while remaining accountable at every step?
Based on
