Source: Financial News Agency Author: Xia Junxiong
① The US government issued new regulations on artificial intelligence (AI), requiring federal agencies to be more careful and transparent when using AI, and stated that with the rapid development of AI, these rules are needed to protect the public; ② US federal agencies must take steps to comply with these new requirements by December 1 of this year.
On Thursday (March 28) local time, the US government issued new regulations on artificial intelligence (AI), requiring federal agencies to be more careful and transparent when using AI, and stated that with the rapid development of AI, these rules are needed to protect the public. But the new policy also has provisions encouraging government agencies to innovate AI where the technology can be used in the public interest.
US Vice President Harris said at a press conference that the government plans to make these policies a model for the world in regulating AI. She said the US will continue to call on other countries to follow their steps and put the public interest first when governments use AI.
The new US AI regulations were issued by the White House Office of Management and Budget. They require the government to be more transparent about how to use AI, and also call on federal agencies to develop more AI technology. This policy shows that the US government is trying to balance AI risk reduction with the use of AI tools to address existential threats such as climate change and disease.
According to new regulations issued by the White House Office of Management and Budget, US federal agencies must first ensure that the AI tools they use do not pose a risk to Americans. For example, if the Department of Veterans Affairs is to use AI in its hospitals, it must verify that the technology doesn't give racially biased diagnoses.
Second, the federal government needs to improve the transparency of AI systems. As long as the published information does not pose a threat to the public or government, each agency must publish the AI models, data, and code it owns. Additionally, federal agencies must publicly report annually on how they use AI, the potential risks posed by the system, and how to mitigate those risks.
The new rules also require federal agencies to strengthen staff expertise in AI and require each agency to appoint a chief AI officer to oversee all AI used within the agency. The focus of this position is to promote AI innovation, while also being aware of its dangers.
US federal agencies must take steps to comply with these new requirements by December 1 of this year. If an agency cannot verify that the AI system it uses complies with the new regulations, it must stop using it or prove that it is reasonable to continue using it.
edit/lambor