28.1 C
Delhi
Friday, July 1, 2022

Mattis: Putin Goes to Bed at Night ‘Fearful’

https://ift.tt/XLi6k1y U.S. Defense Secretary Jim Mattis on...

Following Indonesia’s Diplomatic Blitz, Western Leaders Say They’ll Attend G-20 Bali Summit

https://ift.tt/bVCTmiM leaders signaled confirmation of their attendance...

Ex-Google AI ethics chief: Boost worker power to curb harmful AI

Governments need to curb the power of tech companies and increase the power of workers in order to safeguard people from unsafe uses of artificial intelligence, Google’s former AI ethics chief Timnit Gebru told a hearing at the European Parliament on Tuesday.

“When asked what regulations need to be in place,” she said, “I say the number one thing that would safeguard us from unsafe uses of AI is curbing the power of the companies who develop it, and increasing the power of those who speak up against not only the harms of AI but the tech companies’ practices.”

Gebru said one good example was the state of California’s Silence No More Act, which makes it illegal for companies to prevent employees from speaking out about harassment or discrimination. “I think this needs to be universal. I think that companies need to have much stronger punishments for violating any of these laws, like the huge aggressive union-busting activities that we see by Amazon,” she continued.

The EU is negotiating the world’s first AI law, the AI Act, which places restrictions on AI uses that are likeliest to cause harm. Italian S&D MEP Brando Benifei, the lead negotiator for the bill, has called for the AI Act to have a redress mechanism for AI harms and more involvement from labor unions.

Gebru is a leading AI ethics researcher, whose work focuses on combating bias. At Google, she co-led the company’s AI ethics team. Gebru said she was fired last year following a disagreement over a paper she wrote on harms around large language models. Google said she resigned, and she has since been working to launch her own independent AI research institute.

Tech companies and the military sector are some of the biggest spenders in AI technology development. Gebru said that in order to “stop the cycle” of AI harms, governments need to invest in other communities building AI technologies.

“Contrary to the Cold War-era rhetoric of the arms race — this is really what stifles innovation, because some people are out there building harmful technology and other people are constantly behind because they’re trying to stop this harmful technology from being built,” Gebru said.

“We can only change when we changed the incentive structure,” she added. “Can we see technology that prioritizes the well-being of citizens, rather than a race to figure out how to kill more people more efficiently, or make the most amount of money for a handful of corporations around the world?”

https://ift.tt/eA8V8J November 30, 2021 at 11:46PM
Melissa Heikkilä

Support Us

Secured by Paypal

Related Stories

Check out other tags:

Most Popular Articles