The US plans to ban China's access to powerful AI models developed by the US
Kyiv • UNN
The United States plans to introduce new rules to restrict Chinese companies' access to advanced artificial intelligence models developed by American tech giants such as Microsoft, OpenAI, and Alphabet for national security reasons.
The U.S. Department of Commerce is considering introducing new rules to make it more difficult for Chinese companies to access advanced artificial intelligence technologies from the United States. Reuters reports this with reference to three of its own sources, UNN reports.
Details
Such regulation will apply, in particular, to closed AI models developed by companies such as Microsoft, OpenAI, and Alphabet. These models, unlike open ones, have closed source code and data on which they are trained, making them unavailable for public analysis.
According to U.S. officials, the flow of algorithms and data to China to train these AI models could pose a threat to U.S. national security.
U.S. government officials and private sector experts fear that hostile forces could use artificial intelligence, which analyzes huge amounts of text and images to summarize information and generate content, to launch aggressive cyberattacks or even create biological weapons.
In addition, one of the news agency's sources said that any new export controls are likely to be aimed not only at China, but also at Russia, North Korea, and Iran.
AddendumAddendum
According to Reuters sources, the US may introduce a threshold level as part of its AI export control measures, as stipulated in President Joseph Biden's October 2023 executive order on AI. This level is based on the amount of computing power required to train the model.