AI Agents Force Enterprise Identity Stack Cleanup

As AI agents move from pilots into production, enterprises are being forced to confront decades of fragmented identity and access management tools. Okta CRO Jon Addison says customers see agentic security as an extension of existing identity work, not a new problem. The pressure to deploy AI fast while maintaining visibility and control is driving consolidation of point solutions that create security vulnerabilities. Some organizations are establishing reference architectures before deploying agents, while others already have shadow agents running and need urgent access management controls.