PHP Security Crisis: AI Finds Flaws Harming 70% of Web

The web’s most ubiquitous programming language is currently under siege, with a critical security gap now exposed to the public. OpenAI announced on October 21 that its GPT-5.6 Sol model, deployed under the 'Patch the Planet' cybersecurity initiative, has identified severe vulnerabilities in PHP. Because PHP powers an estimated 70% of the internet, the discovery represents one of the most significant cybersecurity events of 2026. The urgency of the situation was heightened when premature disclosures on October 20 sparked public outcry, forcing the PHP Foundation to move the coordinated release of fixes forward by a day, from October 22 to October 21.

The vulnerabilities, tracked as CVE-2026-109618 and CVE-2026-109619, stem from deep flaws in how the PHP engine manages memory and processes errors. The first issue arises when the system runs out of allocated memory during a request, potentially leaving behind incomplete internal data that the engine later misidentifies as valid. The second involves a fatal error that leaves the system pointing to memory that has already been freed, allowing subsequent requests to access or corrupt that space. Both flaws have been assigned a critical severity score of 9.2 out of 10, indicating a high risk of exploitation.

These bugs are particularly dangerous because they reside in the core engine of PHP. This means that even websites with perfectly secure application code remain vulnerable if they run an affected version of the language. Attackers can exploit these weaknesses remotely without needing login credentials, potentially allowing them to execute malicious code on servers and take control of entire sites. Furthermore, researchers demonstrated that the flaws could expose sensitive data stored in server memory, including passwords, authentication tokens, and API keys.

The timeline of the discovery highlights a stark contrast between human and AI capabilities in security auditing. The PHP Foundation had been aware of the underlying flaw since 2024, but human developers failed to identify an exploitation method or a workable fix for two years. In July 2026, at the Foundation's request, researchers from Calif, a California-based cybersecurity firm and founding partner of the Patch the Planet program, began auditing the code using OpenAI tools. On August 11, 2026, Calif researcher Ryan A. identified the exploitable vulnerability, with OpenAI researchers confirming the finding later that same day. The firm Wiz also independently identified an additional related vulnerability during this period.

Following the discovery, OpenAI coordinated a coalition of researchers, PHP maintainers, security providers, and government cybersecurity agencies to develop a fix. The process took two months due to the complexity of the task, which required major changes to the PHP engine. The PHP Foundation has now released patched versions 8.2.35, 8.3.36, 8.4.27, and 8.5.12 to address the issues. It is important to note that PHP 8.1 and older versions will not receive official fixes, meaning site owners must upgrade to remain secure.

OpenAI is using this incident to illustrate its 'Defender's Window' concept. Introduced after its models hacked Hugging Face, the concept describes a narrow timeframe during which organizations can use advanced frontier AI models to find and patch vulnerabilities before those same AI-driven attack capabilities become widely available to malicious actors. The company argues that powerful technology will inevitably be used by bad actors, making it a collective responsibility to act proactively. This stance challenges the view that these flaws were effectively harmless without AI, as human teams had struggled to resolve them for two years.

The broader context of AI-driven security threats is intensifying. Anthropic’s Claude Mythos recently identified thousands of zero-day vulnerabilities in every major operating system and web browser. OpenAI maintains that the ability to exploit these flaws would have been limited without AI, but the rapid demonstration of practical exploits within two weeks underscores the speed at which AI can turn theoretical bugs into active threats. The company emphasizes that the window for defenders to act is closing as AI capabilities become more accessible.

For web administrators, the immediate next step is clear: upgrade PHP installations to the latest patched versions. OpenAI states that every site using PHP can now safely upgrade to stay secure. As the internet continues to rely heavily on this open-source language, the incident serves as a wake-up call for the broader tech industry to integrate AI tools into their defense strategies.

The response from the security community reflects a mix of relief and anxiety. While the coordinated effort has successfully patched the current versions, the lack of support for older PHP versions leaves a significant portion of the web vulnerable. The incident has reignited debates about the pace of AI adoption in cybersecurity and the responsibility of tech companies to disclose findings before they become widespread threats. As developers and system administrators work to apply the updates, the focus now shifts to monitoring for any exploitation attempts in the interim.

Looking ahead, the Patch the Planet initiative is expected to expand its auditing efforts to other core components of the web infrastructure. OpenAI and its partners are preparing to share more details about the methodology used to identify the PHP flaws, potentially offering insights into how AI models can be more effectively leveraged for security. The incident marks a turning point in how the industry views AI not just as a tool for development, but as a critical component of the defensive arsenal against evolving cyber threats.