Artificial superintelligent machines are coming, and it won’t be for a while.
Here’s what you need to know.
Bloomberg.com/AI The term artificial superintelligence refers to artificial computers that are more powerful than humans, but also more complicated and less intelligent than humans.
It’s often used to describe the kind of computers we’ll be building in the future.
“It’s not just an academic field,” said Adam Jonas, senior director of engineering and scientific research at Google, at a TEDx talk in February.
“We’re not even sure whether the term artificial intelligence refers to the artificial intelligence that’s going to be around in the next five years.”
Google is using artificial intelligence to create Google Photos, Google Search, and the new Google Maps app.
Google says its AI systems can do everything from make photos to understand speech and navigate the web.
The company also says it’s building a machine-learning system that can learn from photos and text to create products that are both faster and more useful than humans can do.
It’s still early days for the field.
But artificial superintelligences are coming.
They’re being built by the likes of Google and IBM, and they’re coming with greater efficiency and better performance.
The problem is that the machine-intelligence that makes them better is already built into computers.
In February, IBM announced that it had developed an AI system that learns from images, can recognize objects in photos, and even learn to write and read.
AI experts are predicting that by the year 2030, AI will be able to make the decisions we make, from buying cars to shopping online, that humans can’t.
IBM has a lot to gain from AI, but it also has a big problem: It can’t keep up with its artificial super-intelligence competitors.
So, to keep up, IBM and Google have to keep improving.
That means building new machines that are faster and smarter.
This story was produced by The Wall Street Journal and Kaiser Health News.