Recently, a co-worker said AI was so extreme that if put in charge of human happiness, it would strap us to a chair on a heroin drip à la “The Matrix.”
He’s watched too many movies. Mostly, AI simply finds “which of these things is not like the other” in a large set—or handles the opposite task of grouping similar objects. It’s not something to be afraid of anytime soon, really. But his fantasy made me wonder what other things people think is or isn’t coming in the next five years or so.
What you’ll see in five years
Searching videos for dialog: This is already available to some degree. Within the next five years, however, it will be commoditized to the point that you’ll be able to find YouTube clips of your favorite movie taglines even if no one added them to the title or description. The main advances are in signal processing and encoding, as well as speech recognition. But it's close because we’ve been working on it for more than 40 years.
The near ubiquity of Amazon Now: Basically, even if you live in or near a small city like mine (Durham, North Carolina, population 250,000) you’ll soon be able to try on major-brand clothes and get produce delivered within four hours. But it almost definitely won’t be by drone. It will be by poorly paid workers in the gig economy.
Cashierless checkouts: I’m not talking about the self-checkout at the grocery store that takes twice as long because it makes you keep everything on the scale and put everything in the bag with a delay, nor am I talking about Amazon’s somewhat less than successful experiment. I’m talking about something in between. Basically, a retailer only needs to deploy well-placed cameras, Apple/Google Pay, and QR Code scanners (or NFC) to set up an automated store. It would still require a clerk, but probably only one instead of four. Yes, you could probably set up a row of vending machines that take credit cards instead, but how trendy would that be?
The beginning of the end for POTS. The “plain old telephone system” is on its last legs. More and more telecom companies won’t offer POTS—or you’ll be offered something weird, like an old-style phone that connects to a cell tower. Places that are spread out and cater to younger customers may tilt the economics to the point where phone wires aren’t worth it. Then again, my former life insurance company made me fax over a document the other day. I had to find an online service to do that.
Mobile payments (almost) everywhere. Right now, the stupidity of Apple Pay vs. Android Pay vs. Samsung Pay is preventing this from becoming a reality. But soon enough we’ll pay with our cellphones for everything. (We’ll also see a big breach occur, but paying by phone will still be more secure than using a credit card.) The question is when we’ll get a legal framework that protects consumers (hint: not during the Trump administration).
What you won’t see by 2022
Fully autonomous self-driving cars: In the next few years, expect to climb in a self-driving Uber that goes down a well-known route. Think Las Vegas Airport to Las Vegas strip hotel (actually, that route is such a racket, maybe it will come last). This will happen first in the United States where we lack decent, reliable public transportation. The truck lane may also see autonomy ... but fully autonomous self-driving cars won’t arrive in five years. Frankly, our roads aren’t that good, and we’re still far from a car capable of thinking itself out of being stuck on a dirt road. A well-traveled route should probably have a train anyhow.
Drone delivery: This isn’t coming soon. The requirements are too diverse (package size, weight), the navigation is too complex (trees, telephone cables, electric poles), and the legal issues are too daunting. The technology has some maturing to do before a drone can navigate an urban or suburban landscape. Prediction: Amazon quadcopter pads will not be as ubiquitous as mailboxes by 2022.
Lower cable or cell bills: Expect to see more consolidation from a White House kindly disposed toward telecom oligopoly. With cable/telco lackey Ajit Pai running the FCC and “looking out for us” with “common sense” ideas like letting your ISP sell your information and deciding which content you get to see faster, this part of the industry will have no incentive to innovate for the next four years—or if history repeats itself, eight.
A major part of the labor force disrupted by AI: As I say, we’ll still have human drivers for the next many years, despite Uber spending a ton of other people’s money on self-driving cars. Otherwise, the easiest jobs for AI to replace are low-wage positions. Fast food production could have been automated years ago, with no AI, only robots, but it hasn’t happened. The cost savings have to be high enough to make up the investment in a time period short enough to demonstrate gains. Wages have been kept low enough in the United States that investing big bucks to get rid of menial jobs makes questionable financial sense. It will happen eventually, though not as quickly as some people think.
Augmented reality/virtual reality everywhere: Those glasses still make you look stupid. The revived interest in augmented and virtual reality is a fad, like Google Glass was.
Technology may move fast, but not always as fast as people think. Remember when we were going to live on the moon? What do you see coming and not coming as quickly as people think?