I spent a good deal of time yesterday perusing the Economic Report of the President, as smarties like me are wont to do when we’re not blasting through The Dresden Files. While I can’t claim to have read the entire report, one section really popped out for me.
Chapter Five. Technology and Innovation. Subsection: Effect of Robotics on Workers
As a kid growing up in the War Games era, even seeing this header is kind of a weird thrill. We used to have a black and white television with rabbit ears and you could maybe get five channels if you turned the tuning knob juuuuuust right.
When my friends and I played gun fighting games, we used sticks. Just sticks. Bonus points if you found one that was gun-shaped.
Now the President is talkin’bout robots. It’s an exciting time to be alive.
Anyway, I really gobbled up that section of the President’s report.
It’s where I saw this chart.
So, if I’m reading this properly, 83% of workers who make less than 20 dollars per hour will probably be phased out of the workplace. (I feel like I read the date 2030 somewhere but I can’t find it, so I can’t cite it.)
Okay. No big deal, right? How many American workers actually make less than 20 dollars per hour?
Ummmmm…most of them?
According to this data, “19% of workers make less than $12.50/hr, 32% make between $12.50 and $20 per hour.”
That’s 51% of the American workforce. Again, I didn’t read this on some crazy website or in a forwarded email from a blithely racist grandparent. It comes from the President’s OFFICIAL ECONOMIC REPORT.
The US labor force is around 162 million people and change, so that means we’re set to have roughly 51% of those people displaced from their income-earning potential.
That’s 82 MILLION PEOPLE.
Even if these numbers are off by a factor of ten, that amount of labor volatility is concerning. Anyone who has kept an eye on these things shouldn’t be surprised. But when will it happen? How far away is the event horizon on change of this magnitude?
Well, Tesla anticipates autonomous driving cars within the next year and a half. (The cool part is that you can have your car drop you off and it will go find parking, and then you can ‘summon’ it to pick you up when you’re done.) Google is anticipating similar results and the big Detroit automakers have begun to get in on the action. How soon after that will we see actual industry disruption? Five years? Ten? Twenty? How long before our roads are mostly traversed by driverless vehicles?
From the point of view of business owners, the choice seems simple. Why, (without specific legislation to disrupt it) would a company pay for a human truck driver, say, when it could get a flawless, law abiding, instantaneously updating AI driver who doesn’t need sleep or coffee to stay awake or bathroom breaks? Who never has road rage and is never, ever sick at sea?
What will happen to America’s truckers? How many of them are there, anyway?
And that doesn’t include cabbies and uber drivers and bus drivers and livery workers and specialized industry drivers.
Technology is moving faster than ever before and something happened earlier this year which should have been screamed from every mountaintop, but was more like a pebble being thrown into the ocean. It was a huge event, a civilization-changing moment, and people seem only tangentially aware of it.
Why is this such a big deal? Computers win in chess all the time! An AI recently became a grandmaster in 72 hours. Why is ‘Go’ any different?
Because ‘Go’ can’t be brute-force won the way chess can. There are more possibilities in a game of ‘Go’ than atoms in the known universe. And yet a deep learning computer was able to best the greatest human ‘Go’ player on the planet. In the future, when people wondered when things changed, they will look to this moment.
AlphaGo, in Seoul, South Korea, with black and white stones.
If John Connor sends a terminator back from the future, he might just land in South Korea, and AlphaGo might be the primary target.
But learning machines are as ubiquitous in Silicon Valley now as fresh coffee and app developers and they’re only going to get smarter. We’re all poised on the edge of a vast and sweeping transition for humanity, one where the pace of change will make the last 100 years seem like the bronze age.
To learn more, read this article at Wait But Why, maybe my favorite article on the internet, on AI. I borrowed one of his charts for the header picture because I think about it a lot and we should all realize where we are in the technological landscape.
We’re right there. Right on the very edge.
A 2013 study found that 90% of data in the world had been created in the previous two years. And that was three years ago. Thanks to smartphones and the internet, content is being generated by humans at a breakneck pace, and deep learning computers are absorbing all of it. And they’re learning.
AI isn’t without its foibles, as we recently saw with TayTweets, Microsoft’s “Teen AI.”
But AI is coming, and broad workplace restructuring is coming with it. Employers will have the choice between paying human workers or buying robots that never read Facebook during work hours, never have sick kids, don’t need health care, never take vacations, never sleep and never require a paycheck. And some of them can fly.
Which avenue do you think they’ll take?
Soon, we will be outpaced. As sure as the sun rises in the East and sets in the West. It is coming faster than anticipated and more broadly than we had imagined. Artificial Intelligence was once only part of hokey science fiction. Now it’s in our cars and our toasters and dropping bombs on military targets. Our very way of life is poised to change dramatically, and no one quite knows how it will all play out, not even the President’s top advisors.
It’s not something we need to be alarmed about, but it is something we need to think about and plan for and discuss. It sure would be nice if any of the presidential candidates were queried about it as well. Quantum leaps in tech are happening at an increased pace and the rate of discovery will only get faster. The chart from the President’s report only reflects a “probability” of automation, but displacing menial and routine work with thinking machines just doesn’t seem as sci-fi as it once did.
It will all play out, and it will play out sooner than we imagined. The big question is whether the human contingent of the work force will be able to repurpose and retrain at the same pace that technology renders their job obsolete. Based on the rapidity of progress, it seems like that’s going to be a big challenge for humanity.
Truly an exciting time to be alive.