Science fiction movies, like “Metropolis” and the more recent “Ex Machina,” often depict lonely researchers toiling away in secret labs developing super-intelligent, sophisticated robots.
But the director of Facebook’s Artificial Intelligence Research (FAIR) Yann LeCun told Popular Science that this scenario would be impossible:
“The scenario you seen in a Hollywood movie, in which some isolated guy in Alaska comes up with a fully-functional AI system that nobody else is anywhere close to is completely impossible,” LeCun said, “This is one of the biggest, most complicated scientific challenges of our time, and not any single entity, even a big company, can solve it by itself. It has to be a collaborative effort between the entire research and development community.”
While a lone scientist developing an ultra-smart artificial intelligence technology makes for a great movie, it hides the fact that real development in AI is actually because of hard work done by thousand of researchers over the last six decades.
Researchers have been working on perfecting AI that can beat humans at very narrow tasks, but we’re nowhere near the kind of AI that you see in the movies — robots that can do everything an average human can.
These narrowly-useful AI programs are what makes the internet work, silently running behind the curtain of Amazon’s recommendation systems and Facebook’s news feed. But there are still major sticking points — researchers are still figuring out how to make computers accurately explain what they see.
LeCun knows this better than anyone — he has devoted his career to teach computers to see images. Since he joined the FAIR team in 2013, Facebook has been wading further into the artificially intelligent waters and expanding their AI capabilities.
In August, they unveiled M, a personal assistant that can make reservations and book tickets with the initial help of a human working at Facebook. Over time, the program learns from its “AI trainers,” as Facebook calls them, to complete requests on its own, Popular Science reports.
Even as Facebook is training their AI, they work as part of a community — and their work is no secret. LeCun told Popular Science that all of FAIR’s work is open-source, published either on their research site or ArXiv, an open-sourced journal that publishes papers about computer science, mathematics, and physics.
“The research we do, we’re doing it in the open,” LeCun told Popular Science. “Pretty much everything we do is published, a lot of the code we write is open-sourced.”
Business Insider Emails & Alerts
Site highlights each day to your inbox.