IBM Study Warns of Expanding Enterprise Exposure Amid Rapid AI Transformation

• 68% of surveyed executives say meeting data residency and sovereignty requirements across geographies is challenging

• Nearly all (91%) respondents report not fully understanding their AI dependencies across vendors, models, and infrastructure

• Organizations with the most advanced AI control capabilities protect more than half of their operating profit from AI-driven disruptions

A new global study by the IBM Institute for Business Value finds that as enterprises embed AI deeper into core business operations, most surveyed organizations remain locked into AI systems they cannot easily change, reinforcing the growing importance of AI sovereignty to maintain business continuity and performance.

Based on insights from 1,000 senior executives, The Calculus of AI Sovereignty study* reveals that 71% of respondents say switching their primary AI vendor or model would be difficult, highlighting significant operational constraints. Additionally, 68% of surveyed executives say meeting data residency and sovereignty requirements across geographies is challenging, creating complexity in moving AI systems or data across environments. These dynamics point to growing pressure on organizations to strengthen control and oversight as AI adoption and compliance requirements expand.

Also Read: Hyper-Synthetic Data: The Future of Cybersecurity

While the need for control is intensifying, most organizations still lack the visibility required to act on it: 91% of those surveyed say they don’t fully understand their organization’s dependencies across AI vendors, models and infrastructure, limiting the ability to assess risk and plan for disruption. Surveyed leaders report an average of six AI-related disruptions over the past two years, largely driven by vendor services, yet 81% say a seven-day vendor outage would still cause severe or critical disruption, effectively halting operations.

Respondents also cite unexpected changes across the AI ecosystem, including price increases, usage restrictions, model deprecations, and performance degradation. These findings underscore the challenges enterprises face in managing AI dependencies.

Ana Paula Assis, IBM Senior Vice President and Chair, EMEA and APAC, said in the study foreword: “AI has introduced new forms of dependency that evolve faster than traditional governance, procurement, or technology cycles were designed to handle. That is why AI sovereignty has become one of the most defining leadership issues of this moment. The stakes are no longer technical; they are economic. Any loss of control can translate directly into margin pressure, compliance exposure, or outright business disruption.”

According to the study, organizations that design AI systems to adapt data, models and infrastructure as conditions change – a core element of AI sovereignty – are outperforming peers:

• Analysis shows that organizations with the most advanced AI control capabilities see less AI downtime and protect 55% more operating profit from AI-driven disruptions.
• Yet, only a minority of the organizations surveyed (7%) operate at this level, signaling a widening gap between those building adaptable AI systems and those constrained by dependency.
• 72% of surveyed executives say they would accept a 20% cost increase to maintain AI vendors if it improved strategic flexibility.

Most surveyed organizations (73%) describe their AI environments as intentionally multi-‑vendor, yet vendor diversity in practice appears to be driven less by deliberate strategy and more by internal and operational realities¹:

• Independent business unit decisions (69%) and geographic necessity (69%) emerge as the leading drivers.
• Legacy complexity is also widely cited by respondents (57%), reflecting mergers, acquisitions, and historical decisions—common across organizations but less often the primary driver.

Also Read: The AI Architectural Trap: Avoiding One-Way Doors

*Study Methodology

The IBM Institute for Business Value, in collaboration with Oxford Economics, conducted a global survey between February and April 2026 to examine how organizations structure control across the AI stack and how these choices relate to resilience, performance and operating economics. The study is based on responses from 1,000 senior executives responsible for AI, data, technology, or related enterprise capabilities across 16 countries and 17 industries. Additional analysis identified distinct AI control profiles by segmenting organizations based on how they structure control across data, models, infrastructure and applications, and assessing the relationship to resilience, performance and operating economics.

Write to us [⁠wasim.a@demandmediaagency.com] to learn more about our exclusive editorial packages and programmes.