AI finds PHP vulnerability from 2014
A hole that had been sitting in PHP since 2014 got found by AI, and PHP runs something near 70% of websites. The researchers were the good guys, so the PHP team heard about it privately and could start fixing before anyone malicious got there. Microsoft says a China-aligned group has already turned AI agents loose on hundreds of organizations. CrowdStrike counted four separate criminal operators running successful AI agent hacking campaigns between May and August. Same tool, very different manners
Transcript (en)
Around somewhere close to 70% of the entire internet relies on PHP to some degree. And some researchers were using AI to try to find vulnerabilities in PHP, in the code. And they found them. And they quietly let the PHP team, foundation, whatever it may be, know, said, hey, by the way, there's some vulnerabilities in the code. You guys should probably work on this before someone finds it, someone malicious finds it. And then they started to make changes. Microsoft is saying that a China-aligned group used AI agents against hundreds of organizations over the past few months. And then CrowdStrike also said that four separate criminal operators ran successful AI agent hacking campaigns between May and August. We talk about how the Internet is messing with all of us all the time. So if you're into that kind of thing, definitely hit follow.
