Artificial intelligence has revealed a critical security flaw in PHP, a programming language used by a substantial portion of global websites. This vulnerability, named PHP2Shell, allows attackers to potentially execute malicious code on servers, leading to website control and data breaches. The same AI that discovered the flaw, GPT-5.6 Sol, can also accelerate its exploitation by cybercriminals. Ryan Austin of Calif, using this AI, demonstrated that previously difficult-to-exploit weaknesses could be breached rapidly. A joint effort involving PHP developers, Google's Wiz, and OpenAI worked discreetly for weeks to develop fixes, modifying extensive code. This secrecy was to prevent attackers from using AI to create exploits before a patch was ready. Historically, exploiting vulnerabilities required significant expertise, but AI now automates this process, potentially enabling attacks before updates are installed. This PHP flaw is considered the most serious in over a decade, comparable in impact to the 2014 Heartbleed bug. Some suggest AI's complexity may have kept PHP2Shell hidden indefinitely. Although patches were planned for October 22, a leak led to an accelerated release on Wednesday amid reports of attempted attacks and precautionary system shutdowns. The core concern is AI fundamentally altering cyberattack economics and speed, shrinking the time between vulnerability discovery and exploitation.