AI Integration Outpaces Corporate Governance, Report Warns

The report, Building Resilient AI: Managing AI Risk through Governance, Security and Resilience, argues that business leaders must shift their perspective, viewing AI risks as fundamental enterprise resilience issues rather than isolated technology or compliance tasks. As systems gain deeper access to sensitive data and decision-making processes, the authors identify five critical priorities: establishing enterprise-wide accountability, hardening identity controls, managing risks across the entire model lifecycle, vetting third-party vendors, and preparing for multi-layered insurance loss scenarios.

Jonathan Salter, Head of Risk Consulting at AXA XL, notes that the most resilient organizations are those that map exactly where AI is embedded and understand the consequences of failure. While data from the World Economic Forum shows that 64% of organizations now assess AI tools before deployment—a significant jump from 37% a year ago—the report warns that pre-deployment checks are insufficient. Ongoing oversight is required to combat threats like prompt injection, model manipulation, and the rise of shadow AI. Without continuous monitoring, firms remain vulnerable to systemic weaknesses in data governance and supplier oversight that AI effectively amplifies.