AI turns patches into working exploits in 30 minutes, and the 90-day disclosure window is the casualty

2026-05-11 · Source: The Decoder · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Cybersecurity & Data Privacy · Depth: Intermediate, short

Summary

Security expert Himanshu Anand argues that AI language models have rendered the traditional 90-day vulnerability disclosure process obsolete. AI tools enable multiple researchers to find the same security flaws almost simultaneously and allow attackers to reverse-engineer security patches into working exploits in minutes, eliminating the time vendors and administrators previously had to secure systems. Anand cites three real-world examples: eleven simultaneous reports for an online store flaw, a React exploit created in 30 minutes from a patch diff, and the "Copy Fail" Linux kernel vulnerability exploited by Iranian threat actors within days of discovery. He also highlights the "Dirty Frag" Linux flaw, where a negotiated five-day embargo was broken within hours by independent disclosures, leaving distributions without ready patches.

Key takeaway

For CTOs and VPs of Engineering managing cybersecurity, the rapid acceleration of AI-assisted exploit generation means your organization's vulnerability response strategy must fundamentally change. You should mandate immediate deployment of critical patches and push for significantly shorter disclosure timelines from researchers. Waiting for monthly maintenance windows or traditional 90-day disclosures is no longer viable; attackers are already leveraging AI to exploit flaws within hours of a patch release.

Key insights

AI language models accelerate vulnerability discovery and exploit generation, invalidating traditional 90-day disclosure policies.

Principles

• Vulnerabilities are rarely exclusive.
• Patch-to-exploit time is now minutes.
• Embargoes are easily broken by parallel discovery.

Method

Attackers use language models to analyze source code diffs from security patches, rapidly generating functional exploits in minutes, a task that previously took days for experienced reverse engineers.

In practice

• Integrate AI for defensive patch diff analysis.
• Continuously scan dependencies with AI.
• Verify patch efficacy using AI tools.

Topics

• AI-Assisted Exploitation
• Vulnerability Disclosure Policy
• Patch Management
• AI-Accelerated Threats
• Linux Kernel Vulnerabilities

Best for: CTO, VP of Engineering/Data, Executive, AI Security Engineer, Security Engineer, Director of AI/ML

Related on AIssential

• everyone JUST got HACKED... — AI models are rapidly accelerating vulnerability discovery and enabling autonomous, sophisticated cyberattacks.
• Anthropic study shows AI needs hours, not weeks, to build exploits from security patches — LLMs can build software exploits from patches in hours, not weeks, fundamentally changing cybersecurity defense timelines.
• Preparing your security program for AI-accelerated offense — AI accelerates both offensive and defensive cybersecurity, demanding rapid adaptation of security programs.
• Mythos changes the landscape for vulnerability management — AI models like Mythos accelerate vulnerability exploitation, demanding a shift from periodic patching to continuous, automated security responses.
• Attackers Are Exploiting Vulnerabilities 7 Days Before the Patch Exists. Now What? — The patch window is gone; AI-accelerated exploitation now precedes patch availability, demanding a shift to exploitation-based prioritization.
• 😸 A patch wave is coming for your software — AI is accelerating vulnerability discovery, creating a "patch wave" that current security infrastructures cannot handle.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by The Decoder.