OpenAI unveils cybersecurity-focused AI model

April 15, 2026 / 2:43 PM

Sharjah24 - WAM: OpenAI has unveiled GPT-5.4 Cyber, a modified version of its latest flagship model designed specifically for defensive cybersecurity applications. The announcement follows a similar move by rival Anthropic.

Rival model introduced earlier

Anthropic introduced its advanced AI model, Mythos, on April 7 as part of its Glass Wing project. The initiative allows selected organisations to use the Claude Mythos Preview model, which has not yet been publicly released, for defensive cybersecurity purposes.

Capability to detect vulnerabilities

The model has reportedly identified thousands of major vulnerabilities in operating systems, web browsers, and other software, highlighting its potential role in strengthening digital security.

OpenAI said GPT-5.4 Cyber will initially be rolled out on a limited basis to vetted security service providers, organisations, and researchers.

Expanded access programmes

The company also announced plans to expand its Trusted Access for Cyber programme to include thousands of individuals and hundreds of teams working to protect critical software systems.

Enhanced verification tiers

In addition, OpenAI is introducing new levels within its TAC programme, launched in February, offering broader capabilities to users who meet higher verification standards.

Access to advanced capabilities

Users at the highest tier will gain access to GPT-5.4 Cyber, which has fewer restrictions on sensitive cybersecurity tasks such as vulnerability research and analysis.