A cyber emergency dubbed 'Black October' is rippling across the globe. Artificial intelligence just found flaws that left millions of websites open to total hijack.

Government portals and top commercial hubs suddenly vanished offline in the US, UK, and Europe. Operators scrambled to apply emergency fixes before hackers moved in.

The bug resides in PHP, the foundational server language driving WordPress, Wikipedia, and global e-commerce. Up to seven in ten known websites were exposed.

This was not a criminal cyber breach. OpenAI's autonomous research agents uncovered the flaw first, gifting engineers a narrow window to patch the planet.

PHP has powered web servers since the 1990s. For decades, human code auditors missed vulnerabilities that frontier neural models mapped within hours.

Cybersecurity firm Wiz called the flaw one of the most severe in software history. OpenAI confirmed its models discovered the vector before malicious rings weaponized it.

Frontier AI is now aggressively finding zero-day flaws. Does using AI to scan legacy code make the global internet safer or more fragile?

Agencies like the NSA and ENISA remain on high alert. Millions of unpatched private servers remain sitting ducks as automated scanning scripts probe the net.

If you manage a server, update PHP immediately. Autonomous AI security auditing is here, and the race between protectors and exploiters has changed forever.