- by Chris Valance
- technology reporter
Professor Yang Lucan is known as one of the three godfathers of AI and works as a top scientist for Facebook owner Meta
One of the three “godfathers of AI” said AI will not take over the world or destroy jobs forever.
Prof Yann Lucan said some experts’ concerns that AI poses a threat to humanity are “absurd”.
Computers will be smarter than humans, but that’s years away, he said, and “if it turns out to be unsafe, they just don’t build it.”
A British government adviser recently told the BBC that some powerful artificial intelligence may need to be banned.
In 2018, LeCun was awarded the Turing Award for his breakthrough work in AI, along with Jeffrey Hinton and Joshua Bengio, all three of whom have become known as the “Godfathers of AI.”
Professor LeCun is currently working as Chief AI Scientist at Meta, the parent company of Facebook, Instagram and WhatsApp. He disagrees with fellow Godfathers that AI is a risk to humanity.
“Will AI take over the world? No, this is the projection of human nature onto machines,” he said. He added that keeping AI research “closely controlled” would be a big mistake.
LeCun argues that those who worry that AI poses a risk to humans did so because they couldn’t imagine how it could be made safe.
“It’s like asking someone in 1930 how to make a turbojet safe. In 1930 the turbojet hadn’t been invented. It’s the same as not having one.”
“Turbojets have finally become incredibly reliable and safe,” he said, adding that the same will happen with AI.
Meta has a large AI research program, one of its goals is to create intelligent systems with human-like capabilities. Beyond research, the company uses AI to identify harmful social media posts.
At an invited press event, Professor LeCun presented a so-called He talked about his research on Objective Driven AI.
Professor Lecan speaking to the press at the Meta in Paris
He said there is “no doubt” that AI will surpass human intelligence. But researchers are still missing a key concept to reach that level, which will take years, if not decades.
When people express concern about the superhuman machines that may exist in the future, they are referring to artificial general intelligence (AGI). These are systems that can solve a wide range of problems, just like humans.
When AGI existed, he said, scientists feared “within minutes they could turn on a super-intelligent system that could take over the world.” “You know that’s silly, don’t you?”
In response to a question from BBC News, Professor LeCun said there are progressive advances that will probably allow us to have an AI as powerful as a rat’s brain. It won’t take over the world, he argued, “it will still run in a data center somewhere with the switch off.” “And if it turns out it’s not safe, just don’t build it,” he added.
It has been argued that AI could replace many jobs, resulting in some companies suspending hiring for certain positions.
“This is not going to put many people out of work permanently,” Lecan told the BBC. But jobs will change because “we don’t know” what the most prominent jobs will be in 20 years, he said.
Intelligent computers, he said, will bring about “a new renaissance for humanity,” much like the Internet and the printing press.
LeCun spoke on Tuesday ahead of a vote on Europe’s AI law aimed at regulating artificial intelligence.
“They don’t like AI at all, they think it’s too broad, maybe too restrictive,” he said from conversations with European AI startups. However, he said he is not a legal expert.
Professor LeCun said he’s not against regulation, but in his view each application will need its own rules. For example, different rules will apply to AI systems in cars and AI systems that scan medical images.
