January 3, 2020
AI in 2020 and beyond: create a digital replica of your aging parent or yourself
You're racing to the airport, unaware there's a wreck on the highway ahead.
Fortunately, an artificial intelligence-driven system in your vehicle is looking after you. The system automatically checks on your flight—still on schedule—and determines your chances of making it to the gate on time are slim. With your permission, it can proactively book an alternate flight.
"That's the true virtual assistant in the future," says Gartner vice president and fellow David Cearley. "Rather than having conversational interfaces respond to discrete things, it understands the context and can respond to (your) intent."
Much has been said and written about the future of AI, and the role it will play—good and potentially bad—in practically everything consumers and businesses engage in. What pretty much everyone agrees on is that AI will make a profound difference through the next decade and beyond, during which we may see a further blurring between human and machine.
About a year ago, researchers at Pew Research Center and Elon University's Imagining the Internet Center asked the following: "By 2030, do you think it is most likely that advancing AI and related technology systems will enhance human capacities and empower them?"
Of the nearly 1,000 technologists who weighed in, about two-thirds predicted most of us will be better off, with a third thinking otherwise. And most expressed at least some concern over the long-term impact of AI on the "essential elements of being human."
Almost no one disputes the fact that AI will continue to get smarter.
Sriram Raghavan, who heads IBM Research AI, predicts that in 2020 by combining learning with logic, AI will start to develop a "common-sense" reasoning system, to help businesses deploy more conversational automated customer care and technical support tools.
For his part, Jeff Loucks, executive director of Deloitte's Center for Technology, Media & Telecommunications, believes that within 10 years, AI-powered robots may help aging people remain in their homes. And AI embedded in more smart devices will help all of us monitor our health and wellness.
The dark side
Those who worry about the dark side, however, fear that AI will result in data abuse, loss of jobs and an erosion in our ability to think for ourselves.
And AI systems must be trained without prejudice and bias. An NYU study from last year pointed out that the people building out such systems are too white and too male.
Alarm bells have been sounded by some of the most famous names in tech.
Tesla and SpaceX chief Elon Musk has said AI is far more dangerous than nuclear weapons. The late scientist Stephen Hawking warned AI could serve as the "worst event in the history of our civilization" unless humanity is prepared for its possible risks.
For some, "deepfakes" are the immediate concern, especially with the 2020 U.S. presidential election coming up. These rather sophisticated "doctored" videos can make it look like a politician said something outrageous, controversial or out of character.
There's even a market for deepfake porn.
"We are used to Photoshopped photos by now and are sadly also somewhat used to fake news. But we are really not used to being fooled by our own eyes and ears if we see something on a very crisp and clear video," says Lars Buttler, CEO of the AI Foundation, an organization that has developed forensics technology to help identify such fakes.
A virtual AI you
Separate from all that, AI Foundation is developing "personal AIs," kind of avatars of famous people, starting with a digital version of author and spiritual adviser Deepak Chopra. On a phone, tablet or computer, this virtual Chopra can recognize you, respond to your questions and even meditate with you, Buttler says.
By late in the year, though, Buttler believes you'll be able to create your own personal AIs—perhaps of people close to you like your 5-year-old kid or an elderly parent or grandparent. And you'll be able to create a digital replica of yourself that looks, talks and is trained by you.
"A photo tells us what somebody looks like and (is) frozen in time. With a video we also add the elements of the tone of their voice, their mannerisms. But with your own AI, you can literally go back to that point in time and talk to them (or your younger self)," Buttler says.
Such personal AIs might also be used to entertain, teach or, Buttler suggests, become the future AI equivalent of YouTube stars.
Meanwhile, Snapchat is applying deepfakery into a new feature called Cameo. It will let you edit your own face into a customizable video loop or GIF.
We may make use of virtual assistants and personas in all sorts of ways.
Gartner's Cearley envisions a scenario where you're in your kitchen cooking a roast, assisted perhaps by none-other-than late celebrity chef Julia Child. You're not just following some step-by-step video recipe. A virtual Child (or someone like her) is effectively cooking with you, thanks to AI, sensors, and the fact that your oven and other appliances are connected and can talk to one another.
At a critical moment when there's the risk of overcooking the meat, Child might pipe in: "Oh my goodness, it's time to pull it out,'" she might say.
It is an example of how AI can drive "radical simplification," Cearley says. "The computer is not something that sits on my desk. The computer is my home, my car and the environment that I'm working through."
Distributed by Tribune Content Agency, LLC