Brainpower is so yesterday — leave it to AI

Tech titan Elon Musk worries that artificial intelligence will become too powerful for mankind to control.
Tech titan Elon Musk worries that artificial intelligence will become too powerful for mankind to control. AP file photo

Smart people are starting to worry about the brainpower of machines.

A recent report from Harvard said the emergence of artificial intelligence as a weapon poses as much game-changing potential as the airplane and the nuclear bomb.

They worry it could give small countries and terrorists the long-range strike capability of a superpower, the ability to crash our cyber systems and create a channel for fake news that would overwhelm our understanding of what’s real and what’s not.

Elon Musk, the Tesla/Hyperloop/SpaceX dude prone to the grandiose, thinks unchecked artificial intelligence could become an existential threat to mankind.

If we’re not careful, he warns, we’ll end up in a world where humans answer to machines and not the other way around. Last week, he qualified remarks that regulators should never let the AI genie out of its bottle, but he frets nonetheless.

A report released in May calculated that more than half the jobs in the Kansas City area could be automated — a pace quickened as much by artificial intelligence as by robotics — in less than 20 years.

That fits in with a growing range of predictions that the world of work will change more in the next few decades from artificial intelligence than the economy’s been remade by computing power in the last 50 years.

“We may have a third of men between the ages of 25 and 54 not working by the end of this half century,” Clinton-era Treasury secretary Larry Summers has said, “because this is a trend that shows no sign of decelerating. And that’s before we have … seen a single driver replaced (by self-driving vehicles) … not a trucker, not a taxicab driver, not a delivery person.”

So, do we start preemptively killing all the machines? Hold off, for now.

This is also the stuff that helps us, that takes the mundane jobs off our hands so we can concentrate on the interesting stuff.

Machine learning — where computers don’t just calculate, they actually noodle things based on the manmade algorithms driving them — are why the Amazon Echo (often enough) understands your voice. It’s the technology that launches a Google search with a picture instead of search terms. And it’s how self-driving cars teach themselves how to behave at a four-way stop — and why the future of ride-hailing ultimately will go to the fleet with the best AI.

What you might brace for is the coming artificial intelligence arms war. Already, war-gaming is testing AI designed to fool neural networks and to see whether AI-built networks can be so quick and agile that they’ll never be fooled.

The Harvard team suggests that industry and academics have the smartest artificial intelligence, that government is several steps behind. How comfortable you feel about that will correlate with whether you like power in the hands of a corporate board or Congress.

Communities that can harness the technology will fare better than those that don’t. Kansas City got a boost from Google Fiber at the start of this decade, and we’ve seen the emergence of a high-tech start-up scene enthused by it.

But Austin, home to some of the best computer minds in the world at the University of Texas, has Google Fiber, too.

Scott Canon: 816-234-4754, @ScottCanon