From Weeks to Hours: AI and the Death of the Patching Window
In 2020, the average time to weaponize a known vulnerability was measured in weeks. Today, high-profile vulnerabilities are weaponized in hours.
This session breaks down what that shift means in practice and how to operate in a world where AI has industrialized the reproduction of vulnerabilities faster than organizations can begin to patch.
The basic reality is well documented. Motivated and capable adversaries reverse engineer patches, diff to identify what changed, and build exploits in parallel to their target organization's patching process, betting on being faster. This has historically required rare, niche skills.
But it is 2026. That skill is no longer rare. AI is rapidly democratizing vulnerability reproduction, automating significant parts of the work, and shrinking the time required. Just recently, CISA introduced a vulnerability remediation directive that explicitly acknowledges that AI is shrinking the window between vulnerability discovery and exploitation and calls for dramatically faster patching timelines. It no longer feels like a race between a patching process and attackers, because attackers win as soon as the race begins. Deep down, we all knew this was coming.
The industry's answer was always "patch faster." That was never a strategy. Patches introduce risk and change. Businesses must continue to operate, and change management processes exist for good reason.
So the question has to change. Not "can we patch faster?" but "how do we buy time to patch?"
Mitigation is that buffer. Attendees will leave with prescriptive guidance on identifying vulnerabilities that matter, locating exposed systems, applying temporary mitigations to reduce exploitability, and validating that those mitigations are working. None of this replaces patching. It makes patching possible on an enterprise timeline.
By the end of the session, the audience will leave with three key takeaways:
What it means to operate in a world where vulnerabilities are weaponized within hours, not months, and how that changes response timelines How to prioritize vulnerabilities based on real-world exploitability, exposure, and attacker activity, not severity scores alone How to apply and validate mitigations that reduce exploitability and buy the time needed for safe, controlled remediation