The 200,000-square-foot facility arrives at a moment when the company's own research says 94% of enterprises cannot scale AI, and when its own revenue is shrinking.

On July 8, DXC Technology opened a 200,000-square-foot Customer Experience Center in one of Bengaluru's main technology corridors. The building holds an AI Hub, a Cyber Range, forensics labs, a Security Operations Center, and a Network Operations Center. It is among the company's largest global delivery sites.

The more interesting number is not the square footage. It is 94.

That figure comes from DXC's own global study of 2,496 technology decision makers across 22 countries, published in December when the company launched its AdvisoryX consulting arm. Seventy-seven percent of leaders called AI a board-level priority. Sixty-five percent could not build a clear enterprise business case for it. And 94% ran into significant trouble deploying it at scale. The industry has a name for this now: the AI execution gap.

Bengaluru is DXC's answer to it, poured in concrete.

Why a Building, in an Era of Agents

There is an obvious tension here, and it is worth sitting with. DXC spent its June 11 Investor Day telling shareholders that the future of IT services is agent-based rather than labor-based. Management outlined $1 billion to $1.5 billion in cost reductions driven largely by deploying AI internally, positioning the company as what executives called customer zero. Agentic operations in its own security stack have already displaced Tier 1 SOC analyst work.

So why build a large physical hub in India, the geography where the labor-based model was most deeply rooted?

The company's argument is that the hard part of enterprise AI was never the model, rather it is the last mile. Wiring probabilistic systems into deterministic systems of record, the core banking platforms, claims engines, and ERP estates that cannot be allowed to hallucinate. That work is unglamorous, high-context, and difficult to do over video calls with a client's compliance officer.

"Our greatest differentiator is our people," said Ramnath Venkataraman, President, Consulting & Engineering Services at DXC Technology. "Our new Customer Experience Center brings together our exceptional engineering talent in a space designed for deeper collaboration with customers and partners, where we can co-create, engineer, and scale AI-powered solutions that address complex business challenges. By working side by side throughout the innovation journey, we're helping customers move faster from ideas to measurable business outcomes."

The pressure behind the ribbon-cutting

Context matters, and DXC's is not comfortable. Fourth-quarter fiscal 2026 revenue came in at $3.13 billion, down 6.6% on an organic basis. Shares fell roughly 21% on the results. Fiscal 2027 guidance calls for organic revenue to contract another 3% to 5%, with stabilization pushed out toward 2029. Headcount, once above 170,000 at the 2017 merger, sits near 120,000.

Against that backdrop, a flagship center is not a victory lap. It is a repositioning attempt, and a fairly legible one. If services revenue per head is structurally falling because agents are absorbing the ticket-closing work, the only durable margin sits further up the stack, in the design and engineering of AI systems rather than their operation. Experience centers are where that sale gets made.

India is the logical place to make it. DXC employs more than 40,000 people across roughly 18 sites there, its largest delivery footprint anywhere. But India's IT corridor has changed shape. Bengaluru is now the global capital of the captive center, the offshore engineering arm that multinationals build for themselves rather than rent from a systems integrator. Every dollar a US bank spends standing up its own Bengaluru AI team is a dollar it does not spend with DXC. A co-creation lab that puts DXC engineers in the same room as the client's own is a reasonable defense against that drift.

Integration as the Product

What distinguishes the Bengaluru site from the standard innovation-theater playbook is the presence of the operational layer. A Cyber Range and a live SOC in the same building as the ideation studios means a prototype can be attacked, instrumented, and monitored in the same visit. Most experience centers stop at the demo. This one is designed to carry a use case from whiteboard to production monitoring, which is precisely where the 94% falls off.

That thesis aligns with OASIS, DXC's AI-based service delivery platform built with Anthropic's Claude models, now deployed across more than 50 customers in 17 countries.

"Our new Customer Experience Center represents a powerhouse for AI innovation, engineering excellence, and customer collaboration," said Rob Le Busque, President of Asia Pacific & Japan at DXC Technology. "By bringing together our consulting, engineering, and operations expertise in one environment, we're helping customers accelerate AI adoption and create connected enterprises where people work alongside AI agents to engineer and run the systems of record for our customers."

What US Decision-Makers Should Watch

Nearly every large integrator has opened an AI innovation hub in the past thirty-six months yet very few have published evidence that those hubs converted into signed, scaled deployments. The relevant metric is not visitors. It is bookings, and specifically bookings in the regulated verticals where DXC's account depth is real such as core banking, insurance, healthcare, aerospace, and government.

For American CIOs, Bengaluru is now where a meaningful share of their AI roadmap will be engineered. The building is a bet that proximity still beats abstraction. The next several quarters will say whether the bet pays.