On July 8th, reports emerged indicating that OpenAI is scheduled to officially unveil its most sophisticated artificial intelligence model yet, GPT-5.6, this Thursday local time. Earlier, the U.S. government, expressing concerns over the potential misuse of powerful AI systems for reasons related to national security, had asked OpenAI to postpone the release. This led to the model not being rolled out as initially planned last month. Prior to this delay, OpenAI had provided access to GPT-5.6 exclusively to a carefully selected group of partners and had shared comprehensive details about the model with U.S. authorities.

Globally, nations are in a fierce competition to develop state-of-the-art AI models. However, experts caution that such advancements could substantially expedite cyberattacks on complex and outdated technical infrastructures. In light of these concerns, Washington has heightened its oversight of the release of advanced AI models, aiming to pinpoint and alleviate potential risks.

Previously, OpenAI’s rival, Anthropic, also temporarily suspended access to its most cutting-edge AI models, Mythos 5 and Fable 5, for all users. This move came in response to an export control order issued by the U.S. government on June 12th, citing national security reasons. Nevertheless, these restrictions were lifted last week after Anthropic put in place a series of protective measures.

Furthermore, Elon Musk announced today that his company, SpaceXAI, will make its Grok 4.5 model available to the public tomorrow.