OpenAI Unveils GPT-5.6 Sol for Cybersecurity

OpenAI has announced a preview release of GPT-5.6, a new generation of artificial intelligence models specifically designed to strengthen cybersecurity capabilities. The preview is currently available to a limited number of organizations participating in an early access program before a broader rollout.

The GPT-5.6 family consists of three main models:

• GPT-5.6 Sol: The most advanced model, designed for complex tasks, advanced research, and high-level cybersecurity analysis.

• GPT-5.6 Terra: Offers a balanced combination of high performance and computational efficiency.

• GPT-5.6 Luna: Optimized for speed and lower operating costs.

GPT-5.6 Sol has been developed to support cybersecurity researchers and security teams. The model can analyze source code, identify certain security vulnerabilities, suggest potential fixes, assist with debugging, and improve defensive security testing. Its primary goal is to accelerate the work of security professionals while maintaining a high level of safety and reliability.

OpenAI states that GPT-5.6 Sol includes multiple security mechanisms designed to prevent misuse. Requests involving the creation of offensive cyber tools, bypassing security protections, or facilitating cyberattacks are restricted or automatically refused. The model also underwent several weeks of safety testing before its release.

Despite its advanced capabilities, GPT-5.6 Sol is not intended to replace cybersecurity experts. OpenAI emphasizes that the model cannot independently conduct complete attacks against well-secured infrastructure. Instead, its primary role is to assist professionals with research, analysis, and defensive security tasks.

The GPT-5.6 Sol preview is currently available only to a limited number of selected organizations. OpenAI plans to expand access gradually after completing additional safety evaluations and incorporating feedback from participating partners.

With GPT-5.6 Sol, OpenAI continues to advance AI systems that support cybersecurity professionals. By combining powerful analytical capabilities with enhanced safety mechanisms, this new generation of AI has the potential to improve vulnerability detection, strengthen defensive security practices, and contribute to more secure digital systems.