OpenAI staggers GPT 5.6 release as U.S. officials seek partner vetting

Traders Union
All News
Financial News
OpenAI GPT 5.6 rollout

Highlights

OpenAI will grant early GPT 5.6 access to about two dozen partners, pending U.S. government approval, before a broader release.
The U.S. Treasury, Commerce, National Cyber Director, and OSTP requested OpenAI’s restricted rollout due to concerns over cybersecurity threats and sensitive model leakage.
Earlier this month, the Commerce Department imposed export controls on Anthropic’s Fable and Mythos models after Amazon researchers found potential jailbreak risks, prompting a full suspension.

Government review shapes early access plan

The Information first reported that OpenAI was staggering the release of its newest model, with about two dozen partners expected to receive access before a broader launch. People with knowledge of the matter said those organizations would need U.S. government approval before being allowed to use GPT 5.6.

Requests to limit distribution came from the U.S. Treasury Department, the Commerce Department, the White House’s Office of the National Cyber Director and the Office of Science and Technology Policy, according to the people. OpenAI is taking the staggered approach amid concerns that more powerful models could breach existing cyber security defenses, disrupt critical industries or leak to hostile states.

The move fits into the administration’s broader but still improvised effort to balance AI development with risk controls. The White House has outlined plans for a voluntary framework to assess model safety before release, while President Donald Trump and his advisers continue to argue that companies must remain free to innovate.

Anthropic case highlights wider AI security concerns

Recent action involving Anthropic shows how U.S. officials are already intervening on a case-by-case basis when advanced models raise national security concerns. Earlier this month, the Commerce Department imposed export controls on Anthropic, preventing foreign nationals from using its newest models, Fable and Mythos, a step that led the company to suspend the models for all users.

Officials had reviewed Mythos for weeks before its release, with Treasury Secretary Scott Bessent raising concerns about the model’s ability to exploit weaknesses in the global financial system. Several companies across U.S. industry were granted early access so they could test for and patch security risks before a wider launch.

The Commerce Department’s decision was prompted by a report from Amazon researchers who identified a possible jailbreak that could bypass the model’s guardrails and extract software vulnerability information. Negotiations between Anthropic and the U.S. over restoring access to Mythos are continuing, while a separate legal dispute over the company’s refusal to permit unrestricted military use of its technology also remains unresolved.

Our earlier article on management services organisation (MSO) structures in the U.S. legal market explained how law firms and AI-native entrants are using MSOs to attract outside capital for technology and generative AI investment while staying within ownership rules. It also noted that this funding model is accelerating AI deployment in legal services and pushing the industry toward new operating and pricing models as competition intensifies.

This material may contain third-party opinions, none of the data and information on this webpage constitutes investment advice according to our Disclaimer. While we adhere to strict Editorial Integrity, this post may contain references to products from our partners.