A wave of digital outages swept across the globe this week as governments and major organizations scrambled to protect themselves from two critical security holes in PHP, a programming language that serves as a foundation for a vast portion of the internet. The situation became so urgent that national security agencies in the United States, Finland, and the Netherlands advised organizations to take their entire websites offline if they could not immediately apply a newly released emergency patch. This precautionary blackout left everything from Finnish public services and the Dutch parliament’s website to various BBC program pages and parts of Wikipedia momentarily inaccessible.

What makes this crisis particularly striking is how the vulnerabilities were uncovered. Researchers at cybersecurity firms Wiz and Calif utilized advanced artificial intelligence tools provided through an OpenAI initiative to spot flaws that had remained hidden for over a decade. Experts describe these gaps as some of the most severe ever detected by AI, noting that they could give hackers a direct path into a server’s private files or databases without requiring any login credentials. Some estimates suggest that up to seventy percent of the web relies on PHP, meaning millions of sites ranging from small blogs to massive corporate portals were suddenly exposed.

The urgency reached a fever pitch when details about the flaw leaked onto social media, forcing the PHP Foundation to accelerate its official warning and repair schedule. Cybersecurity specialists warn that because PHP is often embedded deep within other software like WordPress or hosting platforms, many companies might not even realize their systems are at risk. With reports already surfacing of early attack attempts, experts are urging administrators not only to update their software but to scan their systems for evidence that intruders may have already slipped through the cracks during the window of exposure.