IBM Study Reveals AI Risks: Enterprises Face Control and Dependency Issues

IBM Study Highlights Challenges of AI Sovereignty for Enterprises

On June 17, 2026, IBM released a significant report revealing that many organizations struggle with the complexities of AI sovereignty as they integrate artificial intelligence into their core business operations. The study, which surveyed 1,000 senior executives, indicates that a majority of companies face difficulties in managing their AI dependencies and compliance with data residency requirements across different regions.

Understanding AI Sovereignty and Its Implications

The term “AI sovereignty” refers to an organization’s ability to control its AI systems and data effectively, ensuring compliance with local regulations and maintaining operational performance. According to the findings from the IBM Institute for Business Value, 71% of executives reported that switching their primary AI vendor or model would be challenging. This dependence on specific vendors creates operational constraints that can hinder flexibility and adaptability in rapidly changing market conditions.

Moreover, 68% of the respondents expressed concerns about meeting data residency and sovereignty requirements across various geographies. This challenge complicates the movement of AI systems and data between different environments, increasing the risk of non-compliance and potential disruptions to business operations.

The Lack of Visibility in AI Dependencies

One of the most alarming findings from the study is that 91% of executives do not fully understand their organization’s dependencies on different AI vendors, models, and infrastructure. This lack of visibility severely limits their ability to assess risks associated with these dependencies and plan effectively for potential disruptions. Over the past two years, surveyed leaders reported an average of six AI-related disruptions primarily driven by vendor services.

Despite this awareness, a staggering 81% indicated that even a seven-day outage from a key vendor would lead to severe or critical disruptions in operations. Such vulnerabilities highlight the urgent need for organizations to enhance their oversight and control over AI systems as adoption rates continue to rise.

Economic Stakes Linked to AI Control

Ana Paula Assis, IBM’s Senior Vice President and Chair for EMEA and APAC, emphasized in her foreword to the study that “AI has introduced new forms of dependency that evolve faster than traditional governance.” She pointed out that any loss of control over these systems could translate directly into economic pressures such as margin erosion or compliance issues. As businesses increasingly rely on AI technologies, ensuring sovereignty has become a critical leadership challenge.

The study also revealed that organizations capable of designing adaptable AI systems—those that can modify data handling, models, and infrastructure based on changing conditions—are outperforming their peers significantly. For instance:

• Companies with advanced AI control capabilities experience less downtime related to AI disruptions.
• These organizations protect 55% more operating profit from potential AI-driven disruptions compared to those lacking such capabilities.
• However, only 7% of respondents operate at this advanced level, indicating a substantial gap between those who can adapt their systems effectively and those who remain constrained by existing dependencies.
• A notable 72% stated they would accept a 20% cost increase to maintain their current AI vendors if it meant improving strategic flexibility.

The Reality Behind Multi-Vendor Strategies