By Paul Sandle and David Shepardson
BLETCHLEY PARK, England/WASHINGTON (Reuters) -The United States will launch a AI safety institute to evaluate known and emerging risks of so-called “frontier” artificial intelligence models, Secretary of Commerce Gina Raimondo said on Wednesday.
“I will almost certainly be calling on many of you in the audience who are in academia and industry to be part of this consortium,” she said in a speech to the AI Safety Summit in Britain.
“We can’t do it alone, the private sector must step up.”
Raimondo added that she would also commit to the U.S. institute establishing a formal partnership with the United Kingdom Safety Institute.
The new effort will be under the National Institute of Standards and Technology (NIST) and lead the U.S. government’s efforts on AI safety, especially for reviewing advanced AI models.
The institute “will facilitate the development of standards for safety, security, and testing of AI models, develop standards for authenticating AI-generated content, and provide testing environments for researchers to evaluate emerging AI risks and address known impacts,” the department said.
President Joe Biden on Monday signed an artificial intelligence executive order, requiring developers of AI systems that pose risks to U.S. national security, the economy, public health or safety to share the results of safety tests with the U.S. government, in line with the Defense Production Act, before they are released to the public.
The order also directs agencies to set standards for that testing and address related chemical, biological, radiological, nuclear, and cybersecurity risks.
(Reporting by Paul Sandle and David Shepardson in Washington; writing by Kate Holton; editing by William James and Deepa Babington)
