Cognizant forms 25,000-person group to deploy OpenAI tools for enterprises

Bridging the AI implementation gap

Cognizant has announced the formation of the OpenAI Frontier Capability Group, a massive organizational shift involving more than 25,000 professionals. While many large enterprises now have the permissions to access frontier AI models, the company notes that many still lack the specific human expertise required to build those tools into existing systems. This new group is designed to provide the specialized labor needed to turn raw AI capabilities into functional business tools.

The company intends to use this workforce to help clients apply OpenAI models to three primary areas: software development, customer service, and general daily operations. By creating a formal organization, Cognizant aims to standardize how these technologies are deployed across various industries. This move follows an internal period of preparation where the company built its own capabilities before offering them to the broader market.

Cognizant's strategy relies on a workforce model that creates Frontier Certified Engineers and Frontier Business Operators. These credentials are issued directly by frontier-model companies, providing a layer of verification for the skills of the staff. This formal certification process is intended to reassure enterprise clients that the people handling their data and systems have undergone rigorous training on specific OpenAI technologies.

Core focus areas for deployment

One of the primary pillars of the new group is AI-native software engineering using Codex. Codex is a model designed to assist engineers as they write, test, and ship code, effectively acting as a sophisticated coding assistant. By focusing on this area, Cognizant helps development teams increase their speed and accuracy when building new software products from the ground up.

The second pillar involves application modernization, specifically using the Cognizant Application Modernizer tool. This tool is powered by OpenAI models and is used across the entire lifecycle of a project, starting with the reverse engineering of legacy mainframe code. The goal is to transform these older systems into cloud-native code, which is more flexible and easier to maintain in modern computing environments.

To ensure safety and accuracy during modernization, Cognizant specifies that human engineers will still review all consequential decisions. This approach combines the speed of AI-driven code generation with the oversight of experienced professionals who have modernized systems for more than 500 clients. This hybrid model is intended to mitigate the risks associated with automating the migration of critical business infrastructure.

Transforming customer operations

The third focus area for the OpenAI Frontier Capability Group is agentic customer operations. This involves applying OpenAI models to sales, service, and support functions to create more autonomous and capable interactions with customers. Cognizant plans to use its own internal experience as a foundation for how it builds these solutions for its corporate clients.

The company has already demonstrated the feasibility of this approach through a project for a global biopharmaceutical company. They co-engineered an OpenAI-powered enterprise assistant that integrates search, content creation, translation, and approvals into a single interface. This assistant has already handled more than 5 million conversations for the client, showcasing the scale at which these models can operate in a professional setting.

The results of that specific pilot were significant for the pharmaceutical company's internal workflows. According to Cognizant, the assistant reduced the time required for content creation by 60%. By automating these repetitive tasks, the company aims to free up human workers to focus on more complex issues while the AI handles the high-volume, routine communications.

Internal training and scale

Cognizant did not build this capability from scratch; it has been developing these skills internally for some time. More than 21,000 of the company's associates have already completed learning tracks on OpenAI technologies. These efforts resulted in the acquisition of more than 40,000 OpenAI badges, which serve as proof of the staff's proficiency with the tools.

The company also uses its own operations as a testing ground for new AI approaches before they are marketed to clients. For example, Cognizant's sales and customer success teams currently use a solution built on OpenAI models to identify and pursue new business opportunities. This internal