Breadcrumb Links
PMN Business
Article Author:
Canadian Press
Tara Deschamps
Published June 19, 2024 • 3-minute read
To save this article, register for free here, or sign in if you have an account.
AI pioneer Geoffrey Hinton stands backstage at the Collision Conference in Toronto, Wednesday, June 19, 2024. The Canadian Press/Chris Young Photo by Chris Young/The Canadian Press
Article Contents
TORONTO — Artificial intelligence pioneer Geoffrey Hinton has become synonymous with apocalyptic warnings predicting the technology could pose an existential threat, but on Wednesday he shone the spotlight on another risk he says is even more imminent for humanity.
“The existential threat is the part I've talked about the most, but it's not the most urgent part,” he told a packed audience at the Collision Technology conference in Toronto.
Ad 2
This advertisement has not loaded yet, but article continues below.
This content is available to subscribers only
Subscribe now to read the latest news from your city and across Canada.
Exclusive articles from Barbara Schechter, Joe O'Connor, Gabriel Friedman, Victoria Wells and more. Daily content from the Financial Times, the world's leading global business publication. One account gives you unlimited online reading of articles from the Financial Post, National Post and 15 other news sites across Canada. The National Post ePaper is an electronic replica of the print edition that you can view, share and comment on on any device. Daily puzzles, including the New York Times crossword.
Subscribe to unlock more articles
Subscribe now to read the latest news from your city and across Canada.
Exclusive articles from Barbara Schechter, Joe O'Connor, Gabriel Friedman, Victoria Wells and more. Daily content from the Financial Times, the world's leading global business publication. One account gives you unlimited online reading of articles from the Financial Post, National Post and 15 other news sites across Canada. The National Post ePaper is an electronic replica of the print edition that you can view, share and comment on on any device. Daily puzzles, including the New York Times crossword.
Register/Sign in to view more articles
To continue reading, please create an account or sign in.
Access articles from across Canada with one account. Share your thoughts in the comments and join the conversation. Enjoy additional articles every month. Receive email updates from your favourite authors.
Sign in or create an account
or
Article Contents
“I think surveillance is something to be worried about, and AI is very good at surveillance.”
For the British-Canadian computer scientist, this is especially worrying because he believes it could help authoritarian regimes maintain power, and that there is little defense against them, in some cases with powers that even the Supreme Court cannot match.
AI's ability to spy on humanity's every move has long been on Hinton's long list of concerns, which he has rattled off in various talks over the past year to encourage the world to think more carefully about AI and consider what guardrails are needed as it explodes into business and beyond.
Among the concerns he listed on Wednesday were the impending rise of lethal autonomous weapons, fake videos, election fraud, cybercrime and unemployment that threatens to widen the gap between rich and poor.
There is also the “alignment problem” because humanity cannot always agree on what is good, and this can have consequences when powerful technologies are put into our hands.
Top Stories
Thank you for your registration!
Article Contents
Ad 3
This advertisement has not loaded yet, but article continues below.
Article Contents
“Some people think dropping 2,000-pound bombs on kids is a good thing, others don't,” he said.
“There are reasons for both, but you can't agree with both.”
But the risk that Hinton cites that has garnered the most attention is a science fiction-esque prediction that AI will pose a threat to combat robots and even to humanity's very existence.
This prediction has divided the tech community, with some saying the existential threat is a distant future possibility and others saying it will never come to fruition because humans can shut down AI at any time.
Hinton left Google, which acquired the neural-networks business he co-founded with two students in 2013, just as concerns about AI were swirling around his head.
“I left Google because I was 75 and wanted some time off and to watch a lot of Netflix,” he joked.
“But as I was leaving Google, I just wanted to sound a warning… that in the long run, these things could get smarter than us and get out of control. This is not science fiction as Aidan Gomez thinks it is. It's reality.”
Gomez, co-founder of Hinton-backed Toronto-based enterprise AI company Cohere, told the Collision audience on Tuesday that he doesn't think the technology is likely to surpass human capabilities anytime soon, and even if it does, he's skeptical that it will create science fiction-like scenarios.
Ad 4
This advertisement has not loaded yet, but article continues below.
Article Contents
When asked how the world can deal with the problem of AI, Hinton acknowledged that “in most cases, we have no idea what AI should do.”
But on the issue of existential threats, he called for the government to conduct large-scale safety experiments because that's “the only thing that has enough power to get big corporations to invest a lot of money.”
When it comes to fake videos and attempts to distort election results, he also has ideas for “building resistance” to AI-generated content that spreads falsehoods — ideas he said he's recently shared with billionaire philanthropists who have asked him for advice.
“My suggestion would be to run a bunch of ads with very convincing fake videos and then at the end say, 'This is a fake video,'” he said.
Hinton's talk was one of the most anticipated at Collision, with organizers expecting 37,832 attendees over the three days, including a record number of female founders.
Hinton's 20-minute appearance was shorter than the nearly hour-long interview he gave at the same event the previous year, but both talks were equally topical.
Hinton's views are influential in the tech world because in 2018 he was awarded the A.M. Turing Award, known as the Nobel Prize of computing, along with Yoshua Bengio and Yann LeCun, who disagree with Hinton's existential threat theory.
This report by The Canadian Press was first published June 19, 2024.
Article Contents
Share this article on social networks
