This story is over 5 years old.

​Why Automation Today Is Like Computers in the 1980s

​It's not hard to see the parallels between early mainframe computers and the first prohibitively expensive, functionally-limited robots

It's not hard to see the parallels between early mainframe computers and the first prohibitively expensive, functionally-limited robots. Surely, robots will see a similar progression over the next few decades, becoming increasingly slick, affordable, smaller, ubiquitous, and capable of just about anything.

A new explainer video about automation, published today by the popular YouTuber C.G.P. Grey, does a great job tackling the question increasingly haunting us dumb humans: What does this trend mean for the workforce, in the future, and right now? It's titled Humans Need Not Apply, and the 15-minute primer is worth a watch.


While the first wave of automation took the form of mechanical muscle, says Grey, the next generation is about artificial intelligence, "mechanical minds," and smart bots that can teach themselves things they aren't pre-programmed to do. That type of vitality will kick the robot economy into high gear.

"General purpose" robots, like Baxter, Rethink Robotics' adaptive industrial robot, cost less than the average human salary and have a nearly endlessly versatile skillset, Grey says. "Baxter today is the computer of the 1980s," says the video. "He's not the apex, but the beginning."

In other words, as robots get more useful, demand goes up, prices go down, and they get smaller, more user-friendly and start to proliferate, just like computers have over the last 30 years.

This photo of the now-defunct Rover assembly line in the 90s shows how industrial robots have already evolved into something more than one-trick ponies. Image: Spencer Cooper/Flickr

"We think of technological change as the fancy new expensive stuff, but the real change comes from last decade stuff getting cheaper and faster," says Grey. "That's what's happening to robots now."

A recent Gartner report on the 2014 technology "hype cycle" put smart robots squarely in the "innovation trigger" category, and gave them about 5-10 years to reach the ultimate "plateau of production." That's not a long time. If you're a teen reading this, it's might just make you a little uneasy about your future job prospects.

Really, most of us didn't need a graph to tell us that. People seem to almost intuitively know (or it could be the decades of robopocalypse sci-fi) that AI, software bots, and smart machines threaten to outpace human ability. The monthly, sometimes weekly reports of another profession falling to automation don't help. (My own profession started employing robots just in the last few months, and this after already being nearly killed off when computers replaced print newspapers.)

The rather depressing video makes a strong case for why just about zero jobs are safe, and it's high time we wise up to that fact. But the general population is still split on the idea. A recent Pew Research report found that experts are split 50/50 as to whether artificial intelligence will create or destroy jobs. It also, more frighteningly, found that unless our current education system drastically changes, the rise of robot workers is certain to lead not to a post-scarcity age of mass leisure, but "to an increase in income inequality, a continued hollowing out of the middle class, and even riots, social unrest, and/or the creation of a permanent, unemployable 'underclass.'"

The Reddit discussion thread about Grey's explainer video was similarly divided between dystopian and utopian future-predictors. But the video didn't bother to speculate as to whether we should welcome or dread automation, only that it's inevitable, and indeed is already here in nearly every sector of the economy.

"This is an economic revolution," it says. "You may think we've been here before, but we haven't. This time is different."