OpenAI Unveils GPT-5.6 Sol, a Cybersecurity-Focused AI Model with Government Oversight

OpenAI has announced the preview of GPT-5.6 Sol, its most advanced AI model designed for cybersecurity applications. The model features enhanced vulnerability research capabilities and introduces a new safety stack to mitigate risks. Its release is controlled and supervised by U.S. government authorities, ensuring oversight during deployment. The announcement targets cybersecurity defenders evaluating adoption strategies but does not specify a public release date or technical metrics like model parameters. No CVE IDs or specific attack vectors addressed by the model are mentioned in the report.