NTT DATA unveils global Nvidia-powered AI factories

NTT DATA has launched a global offering it calls "enterprise AI factories", built on Nvidia technology, as large organisations look for a repeatable path from pilot projects to production AI deployments.

The services group is positioning the model as a standard operating approach that combines infrastructure, software and governance in a single environment. The aim is to reduce the fragmentation that can arise when AI projects sit in separate teams and technology stacks.

Operating model

NTT DATA describes an enterprise AI factory as a full-stack, domain-specific environment that brings together data, workflows, infrastructure and governance. The model is intended for use across industries, with configurations tailored to requirements in areas such as healthcare and manufacturing.

The offering uses Nvidia's GPU-accelerated computing and high-performance networking, alongside Nvidia AI Enterprise software. Customers can apply the same approach across public cloud, on-premises data centres and edge environments.

Many organisations have struggled to industrialise AI, often running into inconsistent data access, limited controls, and the effort required to move models from development into operational systems. Service providers and technology vendors have responded with packaged platforms, reference architectures and managed services intended to make deployments more predictable and repeatable.

The approach also reflects a shift in enterprise AI from single-use models to systems that operate across business processes. NTT DATA is emphasising "agentic AI"-software agents that can take actions across systems based on goals and context, rather than only generating responses to prompts.

Nvidia stack

As part of the launch, NTT DATA has expanded its AI solutions portfolio by integrating Nvidia AI Enterprise components, including Nvidia NeMo and Nvidia NIM microservices.

NeMo is positioned as a modular software suite for building, customising and managing AI systems on GPU infrastructure. NIM microservices provide pre-built containers and application programming interfaces for deploying AI applications. NTT DATA said these components form a full-stack platform for building and deploying agentic AI systems.

Abhijit Dubey, CEO of NTT DATA, said customers want AI embedded into core operations with measurable outcomes.

"Visionary enterprises are redesigning core workflows end to end with AI, and they need trusted partners working in unison to achieve transformative and measurable results," said Abhijit Dubey, CEO, NTT DATA, Inc. "By integrating NVIDIA technologies into our enterprise AI factories, we're giving clients a powerful, standardized and secure environment to adopt agentic AI with measurable returns from the start."

Sector examples

NTT DATA pointed to early deployments using Nvidia infrastructure in sector-specific settings. In healthcare, it said a cancer research hospital is working with NTT DATA and Dell using Nvidia HGX platforms for radiology analysis and model evaluation, linked to clinical research and diagnostic workflows.

In automotive manufacturing, it said a global automotive supplier is working with NTT DATA on smart-factory modernisation using "GPU as a Service" based on Nvidia AI infrastructure. The supplier reduced production set-up time from months to days after validating workloads on bare metal and scaling them through an AI-factory architecture.

In technology manufacturing, NTT DATA said a US-based advanced manufacturing company is using Nvidia-accelerated simulation and 3D visualisation to virtually validate a next-generation battery production line. The approach models material flow, automation logic and production scenarios before physical deployment.

Partner positions

Yutaka Sasaki, President and CEO of NTT DATA Group, framed the offering as a response to organisations moving beyond limited deployments towards platforms that combine generative AI with agents operating inside enterprise systems.

"Organisations worldwide are moving from isolated model adoption to intelligent AI solutions and platforms, often complementing GenAI with agents that reason, act and adapt within enterprise systems," said Yutaka Sasaki, President and CEO, NTT DATA Group. "Embedding NVIDIA technologies into our platforms accelerates innovation while giving clients the performance, control and compliance they require."

Nvidia used the announcement to highlight the challenge of scaling AI from proof-of-concept work into sustained operations.

"Enterprises are now seeking robust, scalable platforms that can successfully transition their AI initiatives from pilot projects to full-scale production," said John Fanelli, Vice President, Enterprise Software, NVIDIA. "NTT DATA's AI factory offerings, built on the NVIDIA full-stack platform, provide clients with the domain-specific solutions needed to confidently achieve production-grade enterprise AI at scale."

NTT DATA said it participates in several parts of Nvidia's partner ecosystem, including Nvidia's Solution Provider, Cloud Partner and Global System Integrator Partner Network tracks. It said this covers advisory services, deployments and ongoing operations for customers adopting the AI factory model.

The enterprise AI factories will be available globally and applied in sectors including healthcare, manufacturing and automotive, with further domain-specific implementations expected as customers expand production deployments.