OpenAI will begin the broader public rollout of its GPT-5.6 model family on Thursday, 9 July, following a delay requested by the US government over national security concerns.
The models had initially been made available only to a small group of vetted partners while additional testing and consultations took place between the company and US officials.
US restrictions lifted
The US Department of Commerce approved the wider release of GPT-5.6 after further government testing, Axios reported, citing a person familiar with the process.
OpenAI confirmed that GPT-5.6 Sol, Terra and Luna will launch publicly on Thursday, while access to the limited preview is already being expanded internationally.
The White House and the Department of Commerce did not immediately respond to Reuters’ requests for comment. The government’s approval has therefore been reported by Axios but has not been publicly confirmed by the US authorities.
OpenAI had introduced the GPT-5.6 series on 26 June but restricted access to selected companies and organisations whose participation was disclosed to the US government.
The company said at the time that the phased release would allow for further safety testing and coordination before the models became more widely available.
Three models for different uses
GPT-5.6 is a family comprising three models designed for different levels of performance, speed and cost.
GPT-5.6 Sol is OpenAI’s flagship and most capable model. The company says it offers improvements in software development, scientific research, computer use and cybersecurity.
GPT-5.6 Terra is positioned as a lower-cost model for everyday professional work, while GPT-5.6 Luna is the fastest and most economical option in the series.
OpenAI has described the safeguards accompanying the new family as its most extensive to date, with additional controls intended to prevent the models from being used for harmful cyber activity.
Growing scrutiny of advanced AI
The delayed rollout reflects growing scrutiny in Washington over the potential national security risks posed by increasingly capable artificial intelligence systems.
US officials are examining whether advanced models could be exploited by foreign governments or other actors for cyberattacks, intelligence gathering or the identification of software vulnerabilities.
The Trump administration has also been working with leading AI companies on a framework for assessing the capabilities and risks of advanced models before they are released more widely.
Sources: Reuters, Axios, OpenAI.
