OpenAI's ChatGPT-5.6 is under federal restrictions akin to those imposed on Anthropic’s Mythos model. According to a source, Washington has cautioned OpenAI against releasing the model without first obtaining approval. This follows a similar situation with Anthropic, where the U.S. government initially suspended access to Mythos and Fable before granting limited redeployment to critical infrastructure organizations.

The U.S. government has taken a cautious approach to advanced AI models, ensuring they are not misused. Anthropic received permission to redeploy Mythos 5 to a select group of U.S. organizations involved in defending critical infrastructure. This decision highlights the government's desire to control the deployment of powerful AI tools that could pose risks if used improperly.

The restrictions on ChatGPT-5.6 mirror those placed on Anthropic’s Mythos model, indicating a broader regulatory strategy. The government has allowed Anthropic to restore access to Mythos 5 for over 100 institutions, including major corporations and government agencies. However, the process has been tightly controlled, with the government maintaining oversight over how and to whom the model is deployed.

Such regulatory actions may increase compliance costs for AI developers and delay product launches. Companies may face additional layers of scrutiny, leading to extended approval timelines and increased operational complexity. These measures could also influence market competition, as developers may prioritize models that align with regulatory expectations to avoid delays.

The federal government's intervention underscores the growing importance of governance in AI development. As models like ChatGPT-5.6 and Mythos 5 become more powerful, regulators are taking a more active role in ensuring their safe and responsible use. This trend is likely to continue as governments seek to balance innovation with security and ethical concerns.