Enterprises aren’t struggling to adopt AI, but they are struggling to scale it, according to a report from Tata Communications and Bloomberg Media Studios.

AI has become a universal corporate mandate, but a global study from Tata and Bloomberg reveals a sharper question taking shape inside enterprises: AI investment is no longer in doubt, but the systems beneath it may not be built to carry it, at scale.

According to the report, three in four enterprise leaders (77%) now treat AI as a board-level priority. Yet 65% are still operating on legacy or developing infrastructure not designed for the data intensity and integration demands of enterprise AI. Just 29% say their infrastructure can scale with evolving business demands, a critical gap, given that AI workloads do not rise in a smooth line. They surge, shift across environments and place pressure on the weakest parts of the system.

The study, which surveyed 501 senior executives across North America, Europe and Asia at enterprises with revenues above $500m, identifies five reinforcing systems or loops that determine whether AI investment compounds in value or plateaus over time. The loops span foundation (infrastructure modernisation), integration (interoperability across systems), skills (capability distribution), governance (decision velocity) and RoI (visibility of value). Enterprises can generate isolated gains even when a single loop is under strain. Lasting performance, however, depends on alignment across all five: when the loops reinforce one another, progress accelerates and advantages compound; when any one stalls, constraints spread and momentum weakens.

The research finds pressure points emerging across each of the five loops:

• Foundation: Modernisation is uneven. Fewer than half of enterprises report fully modernised network connectivity, hybrid deployment flexibility or data architecture. Enterprises with advanced infrastructure are nearly twice as likely to report realising high business value from AI as those operating on legacy systems.
• Integration: 28% of leaders cite difficulty integrating AI with legacy systems as a primary roadblock to value, while 38% say integration concerns contribute to delays in approval and procurement cycles. Two-thirds (67%) view the seamless blending of digital automation and human interaction across channels as critical to AI execution.
• Skills: 30% of enterprises cite skill gaps and a shortage of specialised talent as a primary barrier to realising AI value. The pressure intensifies with scale; 45% of enterprises with revenues above $5bn cite the skills gap, well above the study average.
• Governance: 42% of enterprises identify security and compliance reviews as the largest source of approval delays, followed by integration concerns (38%) and procurement complexity (38%). As stakeholder committees grow for higher-value investments, governance risks becoming a brake on scale rather than a discipline that supports it.
• RoI: Nine in ten enterprises see some value from modernisation initiatives, yet more than six in ten say they have not reached optimal outcomes. Part of the challenge is visibility; when AI, infrastructure and security are each tracked in isolation, the broader impact across the business stays hidden. Value looks contained inside individual programmes and reinvestment follows that narrower signal.

“AI has become one of the defining business priorities of our time, but the real differentiator is no longer AI itself, it’s the infrastructure and integration that enable AI to deliver value at scale,” said Sumeet Walia, president of Tata Communications. “Our research shows that while enterprise ambition is accelerating, readiness remains uneven. The organisations that will lead in the years ahead are those investing in the foundations that connect people, systems, data and intelligence across the enterprise.”

To explore the findings of the Building Durable AI Advantage report, visit www.tatacommunications.com/hubfs/library/digital-fabric/documents/durable-ai-advantage-bloomberg-report.pdf.

The report is based on a survey of 501 executives and leaders directly involved in or influencing telecom infrastructure and procurement decisions. Fieldwork was conducted between December 2025 and January 2026 across markets including the USA, UK, Germany, France, Singapore, Hong Kong, China, India, Benelux and Nordic regions. Some 65% of respondents were C-suite executives and 35% VP-level or above.