NTT DATA has announced the global expansion of its AI Factories, a network of hands-on environments where organisations can test how priority artificial intelligence use cases can be deployed, secured, governed and operated at scale. The announcement, made on October 29, 2026, is aimed at a familiar problem in enterprise technology: many companies can run an AI pilot, but far fewer are confident about moving that pilot into full production.
The factories are not simply demonstration rooms. They are working environments where business and technology leaders can evaluate AI workloads against real-world requirements, including infrastructure architecture, performance, workload placement and cost. In other words, they give organisations a place to answer the practical questions that usually arrive too late, after a system has already been bought and deployed.
That focus matters because, according to NTT DATA’s market research, enterprise leaders want greater confidence in performance and business value before they scale their AI investments. For many organisations, the move from pilots to production is held back by infrastructure bottlenecks, governance concerns, security requirements and uncertainty about operational costs and performance. The factories are designed to help clients build confidence before they scale.
Key Takeaways
- NTT DATA has expanded its global AI Factory network, giving organisations hands-on environments to test AI use cases at scale.
- The facilities help leaders evaluate infrastructure, performance, workload placement, cost, security and governance before scaling AI investments.
- New sites in Riyadh, Paris and Singapore join the inaugural factory in Plano, Texas, alongside a Munich facility that opened on October 7.
Why pilots stall before production
A pilot usually succeeds in a controlled setting. It has a small data set, a limited group of users and a team watching it closely. Production is different. It has to work reliably for many users, meet security and governance rules, cope with changing workloads and still make financial sense. That gap is where many AI projects slow down, even when the underlying technology works.
NTT DATA’s research points to the same pressures. Infrastructure bottlenecks can limit how quickly a system responds. Governance questions can delay approval. Security requirements can change how data is handled. And uncertainty about operational costs and performance can make it hard for leaders to approve a larger rollout. Testing those issues in a production-grade environment, before a final investment decision, is the core idea behind the AI Factory approach.
To support that testing, NTT DATA AI Factories combine the company’s full-stack AI capabilities with technologies from leading ecosystem partners, including Cisco and NVIDIA. The environments are supported by technologies from NTT DATA’s partner ecosystem, including Cisco Secure AI Factory with NVIDIA, which provides a secure and scalable infrastructure foundation for enterprise AI.
The factories are designed to help clients build confidence before they scale.
What the factories actually do
The facilities demonstrate use cases across employee productivity, customer experience, intelligent operations, cybersecurity, networking, software development and industry-specific business processes. That breadth is deliberate: AI is no longer confined to a single department, and a factory that only tested one kind of workload would miss how systems interact in a real organisation.
Engagements also include discovery workshops, technical deep dives, proof-of-concept planning and solution design. In practice, that means an organisation can arrive with a priority use case, work through its technical and operational requirements, test it against its own conditions and leave with a clearer plan for production, rather than a general presentation about what AI could do.
- Hands-on testing of how priority AI use cases can be deployed, secured, governed and operated at scale.
- Evaluation of infrastructure architecture, performance, workload placement and cost against real-world requirements.
- Use cases spanning employee productivity, customer experience, intelligent operations, cybersecurity, networking and software development.
- Discovery workshops, technical deep dives, proof-of-concept planning and solution design.
- Technology from partners including Cisco and NVIDIA, including Cisco Secure AI Factory with NVIDIA.
That approach connects directly to a wider debate in enterprise AI, about whether companies are buying AI faster than they are making it pay. Organisations that test workloads, costs and governance in advance are better placed to judge whether a system will deliver measurable business value, rather than simply adding another pilot to a growing list. The pressure to turn AI spending into results is one reason a production-grade test environment can be more useful than another round of demos.
A network across four regions
The Riyadh facility was the first of the newly announced sites to launch, in September 2026, followed by Paris and Singapore. These new sites join the inaugural AI Factory in Plano, Texas. Riyadh serves as NTT DATA’s strategic AI Factory hub for the Middle East and Africa, connecting regional organisations to the company’s expanding global AI Factory ecosystem across North America, Europe and Asia-Pacific.
These four facilities complement the new NTT DATA AI Factory Munich, which opened on October 7 at NTT Global Data Centers’ Munich 2 data centre. The Munich facility enables organisations to test AI systems they own and control against their own use cases in a production-grade environment before making investment decisions.
Together, these facilities form a connected global ecosystem of expertise and capabilities designed to help clients move their AI strategies into production with greater confidence. The geographic spread matters for a practical reason as well: organisations often need to test systems close to their data, their users and their regulatory requirements, rather than relying on a single distant site.
Partners frame the expansion in similar terms. Dilip Kumar, President of Technology Solutions at NTT DATA, Inc., said that AI at scale is ultimately about turning intelligence into measurable business impact, and that organisations need to balance innovation with responsibility and make informed choices about where AI creates value and how it can be deployed sustainably and economically.
Cassie Roach, Global Vice President for AI Infrastructure and Connectivity Partner Sales at Cisco, said scaling AI goes beyond computing, and that enterprises need a secure and resilient infrastructure foundation spanning infrastructure, security, performance and operations, while maintaining control of their data and AI economics. She said Cisco Secure AI Factory with NVIDIA, combined with NTT DATA’s global presence and expertise, gives organisations a strong foundation for moving AI workloads from experimentation into successful production, a challenge that is also visible in the growing scrutiny of major AI deals involving NVIDIA.
Chris Barnard, Vice President for Networking and Infrastructure Services at IDC, said realising AI outcomes starts with building the right AI Factory foundation with strong governance, and that infrastructure, data, security and operations need to work together, supported by the right ecosystem of partners. NTT DATA’s approach, he said, brings those capabilities together to help organisations build operational maturity, manage complexity and turn AI investments into business outcomes, including in fast-moving areas such as the new AI models being tested for security risks.
What to Watch Next
What to watch next is whether organisations using the expanded network move more AI projects from pilots into production, and whether testing in a production-grade environment helps them resolve the infrastructure, governance, security and cost questions that have held back scaling so far. The Riyadh, Paris, Singapore, Plano and Munich facilities will also show whether a connected global ecosystem, rather than a single site, gives clients the confidence they say they need before they scale. Above all, the test for the AI Factory network will be the one its own announcement sets: whether it helps organisations turn AI intelligence into measurable business impact, sustainably and economically, before they commit to production.