OpenAI Commits $1 Billion to Cybersecurity with New Patch the Planet Initiative and Enhances GPT-6 Astra

Inkorr

Source published: 23 Oct 2026, 18:58 · TehranSite published: 23 Oct 2026, 21:34 · Tehran

OpenAI is putting $1 billion into better cybersecurity through its new Patch the Planet program. It is also improving GPT-6 Astra. Photo: НВ — Техно

Introducing the Patch the Planet Program

According to НВ — Техно, OpenAI has announced a new program called Patch the Planet. The program will bring together cybersecurity experts to find and fix weak spots in computer software. OpenAI is setting aside $1 billion to help organizations pay for access to Daybreak, a platform that checks computer systems for security problems. The program will focus especially on organizations with limited money and resources that are responsible for protecting important systems and services people depend on.

Dangerous PHP2Shell Security Flaw Found and Why It Matters

On October 21, researchers found a dangerous security flaw called PHP2Shell in PHP, a technology used to build websites. This flaw could allow hackers to take control of web servers. OpenAI's GPT-5.6 Sol AI model helped discover this problem and another serious security flaw. To help software developers and security teams, OpenAI shared a tool built with Codex on GitHub. The tool checks websites for signs that hackers have used PHP2Shell to break into them. OpenAI warns that the number of very serious security flaws could rise sharply in the coming months because AI-powered hacking tools are becoming more advanced.

OpenAI also announced improvements to its GPT-6 Astra and GPT-5.6 Sol AI models. Both models are now better at changing computer code from one programming language into another. OpenAI described the PHP2Shell incident as one of the most serious cybersecurity breaches in recent memory. Many banks and government agencies still use older computer systems written in programming languages such as COBOL and Fortran. These older systems can make them more exposed to security risks.

OpenAI's decision to launch Patch the Planet shows that the company is becoming more concerned about computer and online security as new threats linked to AI appear. By spending a large amount of money on security checks, OpenAI wants to help organizations with weaker defenses protect the important public services they run. The discovery of security problems like PHP2Shell shows why organizations need to keep checking their systems, watch for new dangers, and update their defenses as technology changes.

Recent cybersecurity problems, including the worrying PHP2Shell flaw, show why organizations need to keep up with improvements in AI security. OpenAI is working to make its security protections stronger at a time when failures to properly monitor and control AI agents have caused serious concerns. To learn more about how these changes affect online security, read about the recent updates to AI safety rules and protections.