OpenAI is embarking on a cautious rollout of its latest AI model series, GPT-5.6, following discussions with the United States government regarding the review of potent AI systems prior to their public release. Initially, the company plans to grant access to a select group of trusted partners, with a broader rollout anticipated in the coming weeks once the review process is completed.
This measured approach stems from dialogues with government officials that centered on AI safety, cybersecurity threats, and the responsible deployment of such advanced technologies. OpenAI has stressed that this preliminary phase is temporary, and broader access is expected after ensuring the models meet safety standards.
The GPT-5.6 series comprises three distinct versions: Sol, Terra, and Luna. Sol represents the most sophisticated model with enhanced performance capabilities, while Terra and Luna are developed to provide more affordable access. OpenAI has highlighted that even the most advanced model, Sol, does not surpass their internal criteria for what constitutes highly dangerous cyber capabilities. Instead, its design is more aligned with identifying and resolving security issues rather than executing cyberattacks.
Despite agreeing to this current limited launch strategy, OpenAI has articulated concerns about the prospect of government approval processes becoming a permanent fixture for AI model releases. The company views this step as a temporary measure aimed at paving the way for establishing a more robust framework for the future deployment of AI technologies.
This strategic move by OpenAI occurs amidst a growing global focus on regulating advanced AI technologies. There is heightened concern among governments over cybersecurity, potential misuse, and the rapid evolution of next-generation AI systems, prompting a closer scrutiny of how these technologies are introduced to the public.