Two PHP Security Flaws Threaten Websites, Google, OpenAI and Amazon Down

JAKARTA - Two serious security flaws have been found in PHP, a programming language widely used to run websites, including WordPress sites. These flaws allow hackers to run harmful code on vulnerable servers from a distance. They may also allow hackers to access private data.

According to CNBC, in a report published on Thursday, October 22, the two flaws, known as PHP2Shell, led several countries to issue security warnings. Google, OpenAI, and Amazon are also working to find weaknesses, detect attacks, and protect systems at risk.

PHP is a free, open-source programming language that runs behind the scenes of websites. It helps manage website content, handle user logins, and connect websites to databases.

Not every website that uses PHP is affected. However, because PHP is so widely used, the threat is a concern for many countries.

On Wednesday, October 21, the United States National Security Agency (NSA) asked organizations that use PHP to prepare security updates right away. The NSA warned that the flaws could give attackers direct access to files and databases on servers.

If an update cannot be installed, the NSA recommends temporarily turning off the affected system until the flaw is fixed.

The US Federal Bureau of Investigation (FBI) also warned that hackers could use the flaws to run code on remote systems. Such attacks could put private data and computer networks at risk.

Cybersecurity agencies in Singapore, the United Kingdom, Australia, Canada, and the European Union issued similar warnings. Companies, government agencies, and other organizations that use PHP systems have been urged to take the threat seriously.

Interestingly, an attempted attack was detected before details of the two security flaws were made public.

Amazon said its cyber threat intelligence team found an attempt to exploit PHP2Shell using MadPot. This system watches for suspicious activity on infrastructure connected to the internet.

The findings suggest that someone tried to exploit the PHP2Shell flaw before it became widely known.

However, Amazon said it did not know how the attackers learned about the two flaws before they were officially announced.

Artificial intelligence (AI) is also being used to find security weaknesses and develop fixes as the threat continues.

Google said its cybersecurity team at Wiz uses an AI model to help find and investigate PHP security flaws.

Google also uses its AI model, Gemini, to assess security weaknesses, test possible attacks, and check whether the fixes that have been prepared actually work.

OpenAI, the company behind ChatGPT, said researchers using its AI model also helped discover the critical security flaw.

Over the past two months, OpenAI said it had led efforts to work with software maintainers, cybersecurity researchers, and security service providers. They coordinated fixes and prepared ways to protect systems before the flaw was publicly announced.

OpenAI considers PHP2Shell one of the most serious security problems in recent times.

The company also warned that AI can be used in two different ways in cybersecurity. It can help researchers find software weaknesses, but cybercriminals may use similar abilities to discover ways to attack systems.

Meanwhile, Cloudflare has added protection against PHP2Shell through a web application firewall (WAF). This security system checks website requests and blocks those that appear harmful.

The protection is intended to lower the risk of attacks while website managers are still installing security updates.

The PHP development team has also released fixes for both flaws. Cybersecurity agencies are asking website managers to check which PHP version they use and update any affected systems immediately.