Making the business case for sovereign AI models as demand accelerates

Geopolitical shifts and national concerns quietly reshape global AI, driving a subtle balance between local control and innovation

As the demand for technological sovereignty grows urgent, international companies are responding with new strategies. 

Organisations implementing artificial intelligence (AI) had so far primarily focused on improving speed, scale and model performance. Their efforts were built on a system where data movement was unrestricted, infrastructure could be managed globally, and information could be kept in a single location.  

As the demand for AI sovereignty grows, this system is crumbling. The latest AI strategies are increasingly defined by complex infrastructure concerns that go beyond technical requirements to include geopolitical and regulatory factors.  

Organisations are demanding data residency and control over the intelligence layer while assessing the impact of varying regional jurisdictions. There is a growing sensitivity to intellectual property safeguards as geopolitical tensions increasingly affect the technology supply chain. 

Consequently, global technology service providers are changing their strategy. Roli Agrawal, chief of CEO office, AI strategy and sales transformation, NTT DATA, a Japanese global information technology services and consulting company, explained how the demand for sovereign AI is changing the technology industry. 

“To serve this demand, NTT DATA is expanding its AI-ready data centre footprint in India — local compute capacity, secure data residency, high-performance connectivity, and sovereign AI infrastructure that lets enterprises and public sector organisations deploy AI while meeting local governance and compliance requirements,” says Agrawal. 

“Sovereign AI demand in India isn’t driven by regulation alone — it’s shaped by four forces converging at once: The need for data residency and regulatory compliance; the requirement to operate at India’s sheer scale; the need to do so at low cost, and the need to serve India’s linguistic and cultural diversity, which global models don't natively capture,” she says. NTT DATA has committed $1.5 billion over the next three years to expand data centre infrastructure and capacity in India. 

Technology sovereignty is increasingly about continuity and control as much as regulation. However, not every market can develop the entire sovereign stack on its own, according to Agrawal. “Resilience doesn’t mean building every layer domestically — that’s unrealistic for almost every economy in the Asia-Pacific [APAC]. Instead, what we are seeing emerge is what I would call hybrid sovereignty: Maintaining strategic control over data, compute and sensitive AI workloads while continuing to leverage the global innovation ecosystem,” Agrawal says. 

“Ultimately, sovereign AI isn’t about isolation; it’s about resilience. It gives organisations the confidence to innovate at scale while retaining control over the data, infrastructure and governance that matter most.”  

NTT DATA’s Global AI Report found that while 94 per cent of APAC organisations consider private and sovereign AI important to their business strategy, only 30 per cent are prioritising sovereign AI in a concrete near-term execution plan. 

The hybrid sovereignty model is likely to spread across economies that have developed domestic data centre infrastructure and have legislated relevant laws. The global sovereign AI market was valued at approximately $40 billion in 2025 and is projected to reach $148.0 billion by 2032, according to a report by MarketsandMarkets Research. The report says that APAC is the fastest-growing region, driven by China and India. More than 60 countries have published formal national AI strategies and many among these are emphasising technological sovereignty.  

Technology service providers will find growth opportunities as the market moves away from an unquestioned reliance on hyperscale providers. The emerging preference for hybrid sovereign cloud models — which integrate domestic computing resources with regulated, auditable environments — is becoming a primary driver for new business.

The writer is an economic analyst and author