“OpenAI” launches a new model to enhance cybersecurity

OpenAI has unveiled its new artificial intelligence model, “GPT-5.4-Cyber,” aimed at strengthening cybersecurity. This comes just one week after its competitor Anthropic introduced its model “Claude Mythos Preview,” signaling intensifying competition in this field.

According to the company’s announcement, the model has been made available to hundreds of users through the “Trusted Access for Cyber” program, with plans for future expansion. It helps organizations identify vulnerabilities through deeper system analysis, enabled by reduced restrictions. Meanwhile, Anthropic’s model stands out for its ability to detect and exploit previously unknown “zero-day vulnerabilities.”

In the same context, U.S. Treasury officials emphasized the need to take these technologies seriously due to the risks of misuse by hacking groups or state-backed actors. This places the global cybersecurity landscape at a turning point, combining opportunities for advanced protection with the scale of potential threats.

Both companies are promoting these models as preventive tools designed to strengthen digital defenses. However, observers believe that the rapid development of these technologies will reshape the concept of cybersecurity worldwide, particularly in terms of efficiently detecting technical weaknesses and countering increasingly sophisticated cyberattack methods that are increasingly powered by artificial intelligence.