The Biden administration is reportedly contemplating the introduction of new regulations aimed at curbing China’s access to advanced AI models, such as those utilized in ChatGPT.
What Happened: According to three sources privy to the matter, the Commerce Department is considering a regulatory push to limit the export of proprietary or closed-source AI models, which keep their software and data confidential, Reuters reported on Wednesday. This move is viewed as a fresh strategy to safeguard U.S. AI from China, succeeding a series of measures over the past two years aimed at blocking the export of sophisticated AI chips to the Asian country.
At present, there are no restrictions on U.S. AI giants such as Microsoft (NASDAQ:MSFT), OpenAI, Alphabet’s (NASDAQ:GOOGL) (NASDAQ:GOOG) Google DeepMind, and rival Anthropic from selling their powerful closed source AI models globally without governmental oversight. The worry is that U.S. adversaries could exploit these models to initiate aggressive cyber attacks or even create potent biological weapons.
The U.S. may employ a threshold based on the amount of computing power required to train a model, as per an AI executive order issued last October, to ...