As expected, the opening keynote of Google’s I/O 2017 developer conference was a doozy. For two and a half hours, CEO Sundar Pichai and a handful of execs rattled off a staggering list of futuristic features and products: A camera that understands what it sees! AI tools a high-schooler can use to help detect cancer! An omniscient, omnipresent virtual assistant! Independent, incredible, immersive virtual reality! To watch the address was to feel like the future had just arrived, all at once, right before you eyes.
Then you go down the list of actual new things, the stuff you can try right now. An Assistant app for iPhone, a way of sending simple email replies without typing them, Google for Jobs. And you realize I/O felt less like Jobsian product reveal and more like a TED talk: good ideas, educated guesses, and impressive research, but precious little practical application. The same could be said for last year’s event, too. Remember that awesome Google Home launch video? You’re still waiting for many of the things it promised. It was a vision for a product, not a product.
Google’s not alone. In many ways, the entire tech world finds itself in limbo. The internet, smartphones, and Facebook conquered the world, and are now ubiquitous. Meanwhile, the next wave of technology lingers just around the corner: Self-driving cars ruling the road, a world will filtered through augmented-reality glasses, and artificial intelligence in every person, place, and thing. All of that and more is definitely coming. Someday. And every day it doesn’t, it feels late.
In fairness, that first wave wasn’t immediate either—the internet stumbled through Web 1.0, the iPhone needed a few years to sink in, and Facebook started as the domain of Harvard kids. And in many ways, that revolution was easy. “The iPhone was this singular thing,” says Mark Rolston, co-founder of design firm Argodesign. “It was a black, five-inch piece of glass, that held in it a lot of what you used to do on a computer.” That made it instantly understandable. You could put it away when you didn’t need it, and it didn’t alter or occlude your perception of the world.
Now Google and others hope to convince people to strap new things to their face, trading this world for theirs. These companies want to know everything about you, so they might connect you and everything you hold dear to their networks. In exchange, they promise some vague improvement to, well, everything.
Google’s announcements made clear that the world isn’t waiting on some still-impossible technology to make all these fantastic things reality. Just the opposite. Neural networks and deep learning let computers identify everything in your photos faster than you can. Amazon Echo and Google Home hear better than most humans. Cars drive themselves. Your VR headset can map the world as you move through it.
And here’s the best part: When all the stuff Google and everyone else keeps promising finally arrives, you won’t even notice. Everything will just be better, simpler, more efficient. “The user experience will be self-evident,” says Brian Blau, an analyst at technology research firm Gartner. “No one has to teach you what to do, because you just already know it.” Your smart lightbulbs won’t feel like a wholly new thing to be understood—they’ll just be lightbulbs. In a few years, when you can shout a question to no one in particular and have a machine reply instantly, it won’t feel remarkable. Artificial intelligence gets better over time, constantly and automatically improving and personalizing. “It’s going to give you more of what you want,” Blau says, “and you won’t have to worry about the next new feature.” It’s all so compelling—and so future-tense.
What Google, and the horde of developers it summoned to the Shoreline Amphitheatre this week, must do now is start inserting this tech into people’s lives in ways people can see and use. Google doesn’t have a technology problem. It has a product problem. To most people, AI remains a buzzword. Ask them to name an example of augmented reality, and those who don’t offer blank stares will mention Pokemon Go. So far everyone talks about AI as a singular, monolithic thing, something synonymous with “computer” when it actually represents an entirely new way of computing, one with implications you cannot fathom. Trouble is, nobody knows what that looks like yet. “The quality of data about any one person, place or thing, in context—me standing here at this time in this place—and what we’re able to computationally do with that moment, has radically changed,” Rolston says. But so what? Google possesses all this data and knows everything about me. How does it make my life better?
Watch any of Google’s upbeat, aspirational product videos, and you’ll see only glimmers of how this interconnected world might one day look if everything goes to plan. You talk to your phone, which talks to your TV, which talks to your light switch, and they all know it’s you talking. You play games with friends in amazing virtual worlds. But the most striking thing about I/O is it felt almost as if Google was pleading with developers to provide the answers Sundar and Larry and Sergei and the rest of them don’t have. All the pieces are out there, but no one’s quite sure how to put them together yet.
Bits of this already are happening, mostly in ways you don’t notice. Google’s search box looks the same as it did in 1997, but the machine-learning that happens the moment you hit enter provides better, more personalized results every time you use it. Photos snapped with your Google Pixel benefit from Google’s incredible computer vision and processing powers. Going forward, Google, Amazon, Microsoft, and everyone else understands that so thoroughly permeating people’s lives means they don’t get to screw up. “There’s a real danger with some of these new technologies,” says Jan Dawson, principal analyst at Jackdaw Research, “that they get overhyped too early and that leads to massive disappointment and disillusionment among consumers and makes it that much harder to get people excited again when the technology is really ready.” This makes taking it slow the right move.
At some point, though, companies like Google must stop talking about how cool the AI- and AR-powered world could be and start showing people how it will work. Google’s keynote made many promises about things technology “could” do and the way you “will” live. What users got in the meantime was a photo-sharing tool, a way to order Panera, and a lot of sci-fi videos. The company’s vision is thrilling in so many ways, but it won’t mean anything until it starts putting that future in your hands.