Organizations most confident in their AI security are being breached at the highest rates, suggesting confidence reflects deployment speed rather than actual protection.

What happened: A FusionAuth survey of 312 technology and security leaders found that 65% of organizations reported a confirmed AI identity breach in the past 12 months, with a further 23% reporting a near miss. Among respondents who rated themselves 'extremely confident' in their AI security posture, 84% had already experienced a confirmed incident—compared with 64% of those 'very confident' and just 17% of those 'not so confident.'

Why it matters: The data suggests that organizations investing heavily in AI deployment and governance—the markers typically used to judge security readiness—are paradoxically experiencing the highest breach rates. As FusionAuth CEO Brian Bell noted, 'confidence appears to be tracking deployment velocity and governance activity, not actual protection.' This means traditional benchmarks for AI security maturity may be unreliable signals of actual safety, creating risk for enterprises betting their controls are sound.

What to watch: The architecture of identity systems emerged as the strongest predictor of breach outcomes. Organizations running multi-tenant SaaS identity platforms reported confirmed incidents at more than twice the rate of self-hosted deployments (83% versus 38%). Meanwhile, 88% of respondents say AI deployment is outpacing their identity and security infrastructure, and 80% report shadow AI—employees connecting AI tools to internal systems without security review.