What 56% Growth in AI-Enabled Breaches Means for Your Security Strategy

AI-enabled breaches growing 56% year-over-year, now one in four of all breaches, is a specific, measured signal that attacker AI adoption is currently outpacing defensive AI adoption at most organizations — worth translating into an actual strategy adjustment, not just noted as a concerning statistic.

What "AI-Enabled" Actually Means for Your Threat Model

This category spans AI-generated phishing content (more convincing, more scalable than manually-written phishing), AI-assisted vulnerability discovery (finding exploitable flaws faster than manual research), and AI-written exploit code (as seen directly in this batch's Siemens PLC advisory). Each of these lowers the skill and time investment required for an attacker to execute a given attack type — meaning threat actors who previously lacked the sophistication for certain attack classes may now have AI-assisted access to them.

Practical Strategy Adjustments

• Reassess phishing-simulation difficulty and training content — AI-generated phishing is measurably more convincing than older template-based phishing, and training calibrated to older attack sophistication may be genuinely outdated
• Prioritize patch velocity for newly-disclosed vulnerabilities specifically, given AI-assisted exploit development can compress the time between disclosure and active exploitation — the traditional assumption of weeks-to-months before a new CVE sees real exploitation is less reliable now
• Evaluate whether your own security tooling has kept pace with defensive AI capability at a rate comparable to attacker AI adoption — a security stack that hasn't meaningfully changed while attacker tooling has measurably accelerated is a real, growing gap, not a static risk