U.S. Commerce Secretary Gina Raimondo on Thursday voiced concerns about the risk of artificial intelligence being used to disrupt the U.S. election during an event focused on a new consortium advising the Commerce Department on AI rules.
“I’ll be honest, I’m very worried,” Raimondo said when asked about a recent AI-generated robocall that imitated the voice of President Joe Biden.
“AI can do amazing things and AI can disrupt our elections here and around the world. We’re already starting to see it,” Raimondo continued. “But it doesn’t mean it has to happen.”
She said the department would work “intensively” on controls like watermarking to help people understand when content is AI-generated.
The Federal Communications Commission on Thursday announced it was making AI-generated voices in robocalls illegal.
Raimondo made the comments during the launch of a new U.S. AI Safety Institute Consortium composed of more than 200 tech companies, academics, public and private researchers, and civil society groups. The consortium will advise the department’s AI Safety Institute, which will develop “red team” testing standards and other guidelines directed by Biden’s executive order on AI last year.
The new consortium includes top tech companies, including OpenAI, Microsoft, Meta, and Google. The consortium, Raimondo said, will ensure that the department’s work is not “done in a bubble.”
Raimondo also said the U.S. must stay ahead of China on artificial intelligence development and work to prevent the technology from falling into the hands of China’s military, in response to a question from Semafor.
“We have to stay ahead of China. We need to invest in, enable innovation so we always lead. We lead now and we need to stay there,” Raimondo said. “Part of the way we do that is by not just investing, but by using our controls to maintain the lead.”
Raimondo stressed that the U.S. does not want to hold back the Chinese economy on AI, but added that “we cannot enable their military apparatus and we always have to stay ahead.”