🇸🇮 Slovenian, 23 October 2026, AI security
Critical PHP Vulnerabilities Exposed by AI
Researchers using AI tools discovered vulnerabilities in PHP, GitLab, Node.js, and Plex, sparking warnings of increased AI-powered attacks.
Experts in cybersecurity have discovered vulnerabilities in several key services, including the programming language PHP, the development platform GitLab, the Node.js environment, and the media server Plex. The vulnerabilities were discovered using AI tools, specifically those developed by Wiz, Calif, and OpenAI. The term ‘Black October’ is being used to describe the growing threat of AI-powered attacks. Matthew Prince, CEO of Cloudflare, which monitors a significant portion of internet traffic, has warned of a rapid increase in attempts to exploit the PHP vulnerability. He noted that within 10 hours of the vulnerability being publicly disclosed, Cloudflare’s network detected thousands of attempts to exploit it, which rose to tens of thousands within 45 minutes, hundreds of thousands within two hours, and millions shortly after. The vulnerability was discovered by researchers who had previously warned about the potential risks of AI’s ability to detect security vulnerabilities. They had cautioned that AI could be used to develop new tools for malicious attacks on information systems. The researchers noted that AI’s ability to process large amounts of code and detect patterns makes it a powerful tool for identifying vulnerabilities, but also increases the risk of new attacks.