 | ||
| Artificial intelligence pioneer Geoffrey Hinton talks at the Thomson Reuters Financial and Risk Summit in Toronto, December 4, 2017. REUTERS/Mark Blinch/File Photo |
LONDON, May 5 (Reuters) - Artificial intelligence might represent a "more urgent" danger to mankind than climate change, AI pioneer Geoffrey Hinton said Reuters in an interview on Friday.
Geoffrey Hinton, widely recognized as one of the "godfathers of AI", recently revealed he had resigned Alphabet (GOOGL.O) after a decade at the corporation, saying he wanted to speak out on the perils of the technology without it impacting his former employer.
Hinton's work is regarded important to the development of modern AI systems. In 1986, he co-authored the key article "Learning representations by back-propagating errors", a milestone in the creation of the neural networks undergirding AI technology. In 2018, he was granted the Turing Award in appreciation of his scientific discoveries.
But he is now among a rising number of industry CEOs openly voicing worry about the prospective danger presented by AI if computers were to reach more intelligence than humans and seize control of the globe.
"I wouldn't wish to discount climate change. I wouldn't want to suggest, 'You shouldn't care about climate change.' That's a significant danger too," Hinton added. "But I think this might end up being more urgent."
He added: "With climate change, it's really straightforward to say what you should do: you simply stop using carbon. If you do that, ultimately everything will be alright. For this it's not at all evident what you should do."
Microsoft-backed (MSFT.O) OpenAI fired the starting gun on a technical arms race in November, when it made AI-powered chatbot ChatGPT open to the public. It immediately became the fastest-growing app in history, hitting 100 million monthly users in two months.
In April, Twitter CEO Elon Musk joined hundreds in signing an open letter advocating for a six-month moratorium in the creation of systems more powerful than OpenAI's recently-launched GPT-4.
Signatories included Stability AI CEO Emad Mostaque, researchers at Alphabet-owned DeepMind, and other AI pioneers Yoshua Bengio and Stuart Russell.
While Hinton shares signatories fear that AI may prove to be an existential danger to humans, he disagrees with suspending development.
“It’s utterly unrealistic,” he remarked. “I'm in the camp that thinks this is an existential risk, and it’s close enough that we ought to be working very hard right now, and putting a lot of resources into figuring out what we can do about it.”
In the European Union, a committee of legislators reacted to the Musk-backed letter, asking on U.S. President Joe Biden to hold a worldwide meeting on the future direction of the technology alongside European Commission President Ursula von der Leyen.
Last week, the committee accepted a significant set of rules targeting generative AI, which would oblige businesses like OpenAI to disclose any copyright content used to train their models.
Meanwhile, Biden conducted conversations with a number of AI firm executives, including Alphabet CEO Sundar Pichai and OpenAI CEO Sam Altman at the White House, pledging a "frank and constructive discussion" on the need for companies to be more open about their systems.
“The tech leaders have the best understanding of it, and the politicians have to be involved,” added Hinton. “It affects us all, so we all have to think about it.”