Companies Fear Severe or Critical Disruption if Their AI Goes Down

A new IBM study warns that companies fear severe or critical disruption if their AI systems go down, highlighting AI sovereignty as a business-continuity issue. Presented by the IBM Institute for Business Value, together with Oxford Economics, the report “The Calculus of AI Sovereignty” was discussed at VivaTech in Paris, where digital sovereignty was among the leading themes alongside AI, cybersecurity and deeptech. IBM surveyed 1,000 senior executives from February to April 2026 across 16 countries and 17 industries. Respondents reported that switching a primary AI vendor or model would be difficult (71%) and that meeting data residency and sovereignty requirements across regions is challenging (68%). Oversight gaps were also noted: 91% said they do not fully understand dependencies across AI vendors, models and infrastructure. On average, companies experienced six AI-related disruptions in two years, and 81% said a seven-day vendor outage would cause severe or critical disruption, stopping operations.