PHP2Shell Zero-Day Exploited Four Days Before Disclosure as AI-Discovered Flaws Hit Internet-Scale PHP

Amazon has confirmed that attackers were exploiting the PHP2Shell vulnerabilities four days before they were disclosed to the public, and that the flaws are now being broadly exploited in the wild. The disclosure came the same day US and UK authorities issued urgent warnings about the bugs, which affect PHP, the programming language underpinning much of the web. The NSA told every organisation running PHP to update immediately or take systems offline, while the UK's National Cyber Security Centre said the flaws could let an attacker seize control of a website's server without a password or any other login credentials.

What the flaws allow

The two vulnerabilities were found with the help of OpenAI and Google models after going undetected for more than a decade, according to the Financial Times. Wiz, the Alphabet-owned security firm that discovered the bugs alongside researchers at Calif, estimates that PHP powers about 70 percent of all websites. The NSA warned that the flaws often provide a direct path to a server's file system or database, which is why its guidance is unusually blunt: patch now, or pull the system offline until it can be fully mitigated. The NCSC is urging UK organisations to take immediate action, and some European governments were reportedly taking sites offline while they patched.

The exploitation timeline

Amazon, the world's largest cloud provider, said on Wednesday that it had identified an attempt to exploit the bug four days before the weakness became public. That gap is the core of the story: the window between first attacker use and defender awareness was not hours but nearly a working week. The PHP maintainers released software updates to patch the vulnerabilities on Wednesday, the same day the warnings went out. For defenders, the practical implication is that any assumption of a clean pre-disclosure period is gone; systems that were exposed before the patch existed may already have been probed or compromised.

AI finds the bugs, and the pressure builds

The discovery is a milestone for AI-assisted security research. OpenAI said it worked with PHP, Calif and Wiz to verify the vulnerabilities and develop the patches, and that to its knowledge frontier AI found these flaws before attackers did. "Given the severity of these flaws and PHP's widespread use, we consider this one of the most impactful security issues in recent memory," the lab said, adding that the same capabilities that uncover flaws with internet-scale impact can also help attackers rapidly develop offensive tools and exploits. The finding lands amid heightened concern about powerful cyber capabilities in new models from Anthropic and OpenAI, which prompted warnings from politicians and financial regulators and led the US government to temporarily limit their rollout earlier this year.

Why PHP matters so much

PHP's reach is largely a consequence of WordPress, the most widely used website-building platform, which is written primarily in PHP and depends on it to function. That makes the blast radius of a server-side flaw enormous: a single vulnerable runtime can sit behind millions of sites, from small blogs to enterprise deployments. The White House said federal agencies were assessing the potential impact on government systems and critical infrastructure, underscoring how quickly a language-level bug becomes a national security question.

What defenders should do now

The immediate priority is the patch released by PHP's maintainers. Where an update cannot be applied at once, the NSA's recommendation is to take the system offline until it can be fully mitigated. Organisations should also treat the four-day exploitation window as a signal to hunt for signs of compromise on any PHP server that was internet-facing before Wednesday, rather than assuming the patch alone closes the incident. The broader lesson is structural: AI is now finding vulnerabilities at a pace and scale that compresses the time between discovery, disclosure and exploitation, and the organisations that fare best will be those whose patching and incident-response processes can keep up with that clock.

Popular Tech News

RECENT POSTS

CATEGORIES

HELP/SUPPORT

SUBSCRIBE US

PopularTechNews.com is a digital news publication that provides comprehensive coverage of global technology trends, artificial intelligence (AI), space exploration, and major corporate shifts within the tech industry. The platform aggregates and reports on breaking developments, curated industry insights, and environmental or scientific stories intersecting with modern technology.

Copyright PopularTechNews. 2027