The "arms race" in the field of artificial intelligence has once again triggered policy concerns. Recently, OpenAI's release strategy for its latest model, GPT-5.6, was forced to be adjusted, serving as a reflection of the shifting regulatory trends within the industry.
According to informed sources, the change in the release schedule directly stems from clear demands by the U.S. government. U.S. regulatory agencies have required OpenAI to adopt a "phased release" approach before widely distributing the model. Specifically, GPT-5.6 will first be made available to a small group of selected partners deemed more reliable. After a period of testing and risk assessment, it will gradually be expanded to a broader user base.
This news was disclosed by Sam Altman, CEO of OpenAI, during an internal staff meeting on Wednesday. It is understood that regulators are increasingly concerned about the performance levels of current cutting-edge AI models and worry that their capabilities may evolve faster than existing control mechanisms can handle.
Notably, regulatory actions are not limited to OpenAI alone. Just two weeks ago, another major AI company, Anthropic PBC, also faced similar regulatory pressure and was forced to suspend access to its cutting-edge model.
