AI-Discovered PHP Flaws Trigger Global Website Shutdowns

A cyber correspondent explains how researchers using advanced AI identified two critical flaws in PHP, a programming language used by much of the web, and why governments and organizations took websites offline. The discussion covers the discovery process, the scope of the disruption, and the risk from attackers attempting to exploit unpatched systems.

■ AI-assisted discovery
- Use of frontier AI models to search for PHP vulnerabilities
- Two critical flaws and a multi-step exploit chain described as beyond practical human discovery

■ Impact and response
- Government and organizational websites taken offline as IT teams scrambled
- Examples including Edinburgh Council, the BBC Doctor Who area, companies in Japan, Belgium, and the Netherlands, and cybersecurity authority sites

■ Threat landscape
- Defenders finding flaws versus attackers reverse-engineering methods with AI
- Risk to websites that have not applied the issued PHP fix

■ Outlook
- Discussion of further spread as more organizations recognize exposure

It is suited to viewers seeking a concise briefing on the incident and its cybersecurity implications, offering context on AI-assisted vulnerability discovery and incident response. Viewers can use it to understand why patching and monitoring are urgent for PHP-based sites.