Key facts
- The U.S. government is in advanced talks with AI companies to create voluntary standards for new model releases.
- The standards would address benchmarks, release timelines, and access to advanced AI models.
- President Donald Trump previously ordered agencies to test advanced AI models before their release.
- Export controls on Anthropic's advanced models were lifted after a national security suspension.
- OpenAI delayed the public launch of GPT-5.6 at the U.S. government's request.
- Google has also been in discussions regarding its advanced coding models.
The U.S. government is reportedly in advanced discussions with artificial intelligence companies to establish voluntary standards for the release of new AI models, with a potential announcement as early as next week, according to the Financial Times.
These standards aim to address concerns about the potential misuse of advanced AI by foreign adversaries, such as China and Russia. The proposed guidelines would set benchmarks for sophisticated models, define release timelines, and clarify access protocols both domestically and internationally.
This initiative follows an executive order issued by President Donald Trump in June, which directed federal agencies to collaborate with leading AI developers to test advanced models prior to their public release and to draft relevant standards. The U.S. Commerce Department recently lifted export controls on Anthropic's advanced Fable and Mythos models, which had been suspended due to national security concerns.
OpenAI also reportedly delayed the full public launch of its GPT-5.6 model at the request of the U.S. government, limiting access to a select group of vetted partners. Google has also engaged in discussions with the government concerning the release of its upcoming advanced coding models, which are expected to possess enhanced cyber capabilities.
Both OpenAI and Anthropic are reportedly preparing for initial public offerings.