PHP Armageddon: OpenAI, Google’s Wiz uncover ancient holes, sparking internet panic

The U.S. National Security Agency issued a critical alert on Wednesday regarding two severe vulnerabilities in PHP, for which patches have already been released by the PHP Foundation.
As a server-side scripting language, PHP powers approximately 70% of global websites—over 33 million sites—and is integral to major content management platforms like WordPress.
The NSA advises that these vulnerabilities are so critical that unpatched sites should be taken offline immediately until updates can be applied, as they allow attackers to compromise server file systems or associated databases.
These vulnerabilities have existed in all PHP versions since 5.0 (released in 2004). Since PHP 8.1 and older versions are no longer supported, administrators are urged to upgrade their servers to the latest stable release, PHP 8.5.12.
The discovery of these long-standing, widespread flaws was driven by artificial intelligence, specifically through initiatives by OpenAI and Google’s Wiz.
Security firm Calif participated in OpenAI’s “Patch the Planet” program to identify open-source vulnerabilities. Using GPT‑5.6 Sol, they uncovered one of the flaws, as detailed in an OpenAI blog post. Meanwhile, Wiz employed its AI agents—Wiz Atlas, Gemini 4 Argon, and GPT-5.6 Sol—to detect multiple PHP vulnerabilities.
Both OpenAI and Wiz adhered to responsible disclosure protocols, collaborating with the PHP Foundation to develop patches before public announcement. They subsequently coordinated to validate findings and test the fixes, involving additional industry partners.
Wiz acknowledged contributions from Automattic, Akamai, AWS, Cloudflare, Fastly, Imperva, Microsoft, and Vercel. Although WordPress relies on PHP, the company clarified that the vulnerability is not specific to its platform; site owners must update their server configurations and coordinate with hosting providers to ensure protection.
OpenAI confirmed that researchers identified these flaws before malicious actors could exploit them, resulting in no known wild exploits. Previously, the complexity of exploiting these bugs deterred hackers, but AI has lowered this barrier. As Wiz noted, large language models can now adapt exploits across various PHP versions, configurations, and memory layouts.
Essentially, this represents a security breakthrough enabled by AI that simultaneously highlights a heightened risk due to AI’s capabilities.
With the vulnerabilities now public, hackers are actively developing exploits. The priority is for millions of websites to apply patches before attacks occur. Consequently, the NSA recommends taking PHP-based sites offline if immediate patching is not feasible.
Related article
OpenAI launches GPT-5.5 Instant as ChatGPT’s new default model
OpenAI launched GPT-5.5 Instant on Monday, replacing GPT-5.3 Instant as the default ChatGPT model. The company states that this update minimizes hallucinations in high-stakes fields like law, medicine, and finance, while preserving the fast response
Nvidia to Derive 25% of Next Year’s Revenue from Funded Labs
Nvidia has deployed nearly US$50 billion into AI labs purchasing its chips, securing commitments for over $500 billion in total.On August 26, CFO Colette Kress informed analysts that these labs, backed by Nvidia’s balance sheet, will account for roug
OpenAI appoints prominent AI doomer to board
Paul Christiano, a prominent AI researcher dedicated to ensuring artificial intelligence remains aligned with human values and under human oversight, has been appointed to the OpenAI Foundation board, the company announced on Wednesday.“I now believe
Related Special Topic Recommendations
Comments (0)
0/500

The U.S. National Security Agency issued a critical alert on Wednesday regarding two severe vulnerabilities in PHP, for which patches have already been released by the PHP Foundation.
As a server-side scripting language, PHP powers approximately 70% of global websites—over 33 million sites—and is integral to major content management platforms like WordPress.
The NSA advises that these vulnerabilities are so critical that unpatched sites should be taken offline immediately until updates can be applied, as they allow attackers to compromise server file systems or associated databases.
These vulnerabilities have existed in all PHP versions since 5.0 (released in 2004). Since PHP 8.1 and older versions are no longer supported, administrators are urged to upgrade their servers to the latest stable release, PHP 8.5.12.
The discovery of these long-standing, widespread flaws was driven by artificial intelligence, specifically through initiatives by OpenAI and Google’s Wiz.
Security firm Calif participated in OpenAI’s “Patch the Planet” program to identify open-source vulnerabilities. Using GPT‑5.6 Sol, they uncovered one of the flaws, as detailed in an OpenAI blog post. Meanwhile, Wiz employed its AI agents—Wiz Atlas, Gemini 4 Argon, and GPT-5.6 Sol—to detect multiple PHP vulnerabilities.
Both OpenAI and Wiz adhered to responsible disclosure protocols, collaborating with the PHP Foundation to develop patches before public announcement. They subsequently coordinated to validate findings and test the fixes, involving additional industry partners.
Wiz acknowledged contributions from Automattic, Akamai, AWS, Cloudflare, Fastly, Imperva, Microsoft, and Vercel. Although WordPress relies on PHP, the company clarified that the vulnerability is not specific to its platform; site owners must update their server configurations and coordinate with hosting providers to ensure protection.
OpenAI confirmed that researchers identified these flaws before malicious actors could exploit them, resulting in no known wild exploits. Previously, the complexity of exploiting these bugs deterred hackers, but AI has lowered this barrier. As Wiz noted, large language models can now adapt exploits across various PHP versions, configurations, and memory layouts.
Essentially, this represents a security breakthrough enabled by AI that simultaneously highlights a heightened risk due to AI’s capabilities.
With the vulnerabilities now public, hackers are actively developing exploits. The priority is for millions of websites to apply patches before attacks occur. Consequently, the NSA recommends taking PHP-based sites offline if immediate patching is not feasible.
OpenAI launches GPT-5.5 Instant as ChatGPT’s new default model
OpenAI launched GPT-5.5 Instant on Monday, replacing GPT-5.3 Instant as the default ChatGPT model. The company states that this update minimizes hallucinations in high-stakes fields like law, medicine, and finance, while preserving the fast response
Nvidia to Derive 25% of Next Year’s Revenue from Funded Labs
Nvidia has deployed nearly US$50 billion into AI labs purchasing its chips, securing commitments for over $500 billion in total.On August 26, CFO Colette Kress informed analysts that these labs, backed by Nvidia’s balance sheet, will account for roug
OpenAI appoints prominent AI doomer to board
Paul Christiano, a prominent AI researcher dedicated to ensuring artificial intelligence remains aligned with human values and under human oversight, has been appointed to the OpenAI Foundation board, the company announced on Wednesday.“I now believe





Home






