A leader in artificial intelligence is urging Congress to regulate it – Valley Public Radio

AILSA CHANG, HOST:
The chairman of the Senate Judiciary Subcommittee, Richard Blumenthal, opened a congressional hearing today on artificial intelligence with these remarks.
(SOUNDBITE OF ARCHIVED RECORDING)
AI-GENERATED VOICE: (As Richard Blumenthal) Too often, we have seen what happens when technology outpaces regulation.
CHANG: But you know what? That is not Senator Blumenthal. This is.
(SOUNDBITE OF ARCHIVED RECORDING)
RICHARD BLUMENTHAL: If you were listening from home, you might have thought that voice was mine and the words from me. But in fact, that voice was not mine. The words were not mine.
CHANG: This marks possibly the first time AI-generated audio was used to deliver an opening statement in Congress. Blumenthal was illustrating how the technology can be used to mimic someone’s voice – one of many concerns raised during today’s hearing. Joining us now is NPR congressional correspondent Claudia Grisales, who has been covering this issue and joins us now. Hey, Claudia.
CLAUDIA GRISALES, BYLINE: Hey, Ailsa.
CHANG: And I trust this is the true Claudia Grisales speaking to me now.
GRISALES: Yes.
(LAUGHTER)
CHANG: Now, much of today’s hearing was focused on Sam Altman, the CEO of OpenAI. That’s the company behind ChatGPT. Altman was testifying for the first time before Congress. What did lawmakers say to him?
GRISALES: So they had a lot of questions surrounding these worries of where this technology could go wrong, especially without guardrails in place. They listed out concerns from the impact on the labor market that AI will replace workers to the spread of election disinformation and also the impact it could have on artists and their work product and whether it could be stolen. The panel’s top Republican, Josh Hawley, said it is up to Congress and Americans to decide if this technology becomes an innovative breakthrough like the printing press or…
(SOUNDBITE OF ARCHIVED RECORDING)
JOSH HAWLEY: Is it going to be more like the atom bomb – huge technological breakthrough, but the consequences – severe, terrible – continue to haunt us to this day?
GRISALES: And we heard Blumenthal at the top there using that AI-generated audio. And it did draw some chuckles and smiles from the audience at that moment. But he said he’s also worried it could be used – this kind of technology – to steal someone’s voice for more nefarious purposes.
CHANG: Absolutely. Well, how did Altman respond to all these concerns from lawmakers?
GRISALES: He said his company was founded on the belief that AI can improve lives, but also, it brings risks in terms of what needs to be managed.
(SOUNDBITE OF ARCHIVED RECORDING)
SAM ALTMAN: I think if this technology goes wrong, it can go quite wrong, and we want to be vocal about that. We want to work with the government to prevent that from happening. But we try to be very clear-eyed about what the downside case is and the work that we have to do to mitigate that.
GRISALES: And he acknowledged that regulation could help mitigate these risks while still allowing AI to grow the U.S. economy. For example, he predicted there will be far greater jobs in the future with AI, arguing that current models like his are a tool and that the current version of ChatGPT will help with tasks with people in their current jobs. But he didn’t rule out that there could be future concerns, and he agreed that AI companies could submit to testing models or new licensing agreements or be overseen by a new government regulatory body. And he also said the most powerful AI models must adhere to safety models, and companies must behave responsibly. And with more powerful systems, the landscape will change.
CHANG: OK, so it’s clear a lot of people care about this. What are the next steps here for Congress?
GRISALES: So lawmakers are really at the earliest stages of trying to develop comprehensive AI legislation, and the U.S. is woefully behind other places, such as the European Union. And if we look at past examples of Congress trying to regulate emerging technology, it’s made a lot of mistakes. So it’s not clear they’ll be able to catch up this time.
CHANG: That is NPR’s Claudia Grisales. Thank you so much, Claudia.
GRISALES: Thank you. Transcript provided by NPR, Copyright NPR.

source