Don’t look for an Android-powered Wall-E anytime soon, but Google’s recent move toward robotics signals a big momentum shift getting under way.
Ready or not, we’re about to enter the age of robotics.
That’s the message underlying Google’s announcement on Wednesday that former Android honcho Andy Rubin would be aiming the company’s next “moon shot” at robots. Google has bought its way into the robotics game, purchasing seven companies to become the foundation of its robotics team, from makers of robotic arms and powered caster wheels to companies specializing in computer vision.
Google might seem like the corporate version of your eccentric rich uncle, throwing money at crazy projects like balloons that can broadcast Wi-Fi, medical record analysis, self-driving cars, and Internet-enabled glasses, but the analyst Ben Schachter says that the “moon shot” moves by Google all make sense.
“People that I’ve spoken with at Google in the past have said if it’s not going to be a $5 billion business, then it’s not worth doing,” Schachter said. That sum represents around an eighth of Google’s current annual revenue, a substantial amount even to Google.
The robotics announcement, he said, highlights that Google’s ambitious side projects are “ready for prime time.”
“Big companies think that these are potentially big markets. It’s the validation of the idea that these are real businesses now,” he said.
If these real businesses can score Google an extra $5 billion per year within a reasonable time frame — say, a decade — that means that Google has laid the foundation for diversifying its portfolio into areas not directly part of software and services.
However, just because Google’s already made a serious investment in robotics doesn’t mean that we’re going to be waited on hand and foot by Data from “Star Trek: The Next Generation” in the next decade. In fact, a science fiction robot as comparatively simple as Wall-E is going to be a stretch for the foreseeable future.
Welcoming our new robot overlords
Ken Goldberg, a University of California at Berkeley professor who works at the multidisciplinary intersection of robotics, social media, and art, thinks that Google’s robotics announcement will change the field.
“I don’t want to jump the gun, but having a company like Google come into this field is going to put a huge boost of momentum into the research side,” he said. The robotics companies that Google has bought so far have brought some intellectual property with them, but Goldberg said he thinks that they’ve been as much about acquiring their talented employees.
It’s more than enough to give robotics a serious dose of funding and development attention, but it’s not yet ready for prime time. “Google sees that this is very viable in some time frame, but it’s not around the corner,” Goldberg said.
“That sets expectations unrealistically high,” he added.
Ruzena Bajcsy, also a UC Berkeley interdisciplinary professor with more than 40 years experience in robotics, said little to temper her enthusiasm for Google’s announcement.
“It’s very exciting that a company that started with software, search engines, came to the realization that the world is not only software. That real problems [it must deal with] are interaction — the physical world with the computational world,” she said.
How will Google-bots make first contact?
The fruits of Google’s labor won’t be seen for a while, but they’re likely to wind up in industries that are already established and lucrative, yet could benefit by some robotic disruption.
“In the next five years, you’ll see a lot of this technology in old-age homes, in hospitals, and maybe in some industrial applications,” said Bajcsy.
Both Bajcsy and Goldberg think that one of the first areas of consumer robot encounters will be in the home care field. Robots will be able to automate some of the more repetitive and dangerous tasks that human caretakers must complete, such as cleaning house or assisting a client with bathing.
“Just like you had to teach the Roomba, you’ll have teach the robot. In simpler environments, [robots] can teach themselves,” Bajcsy said. “More complicated floor plans,” such as houses with two levels, “are impossible right now.”
Still, robots at home make sense, she said, because that’s where the mass market is. Other potential applications include assisting the elderly or infirm when using Google’s self-driving cars, and helping with language processing — from translation to image recognition. Basically, areas where Google already has invested resources are likely to become tasks that its first robots excel at.
Preventing the rise of evil robots
Bajcsy and Goldberg appeared to be representative of robotics experts, as it was difficult to find anybody who thinks that Google’s involvement is going to be bad for the field.
But that doesn’t mean that there aren’t potential problems facing consumer robotics that must be solved. One particular to Google is the double-edged sword of data collection. On the one hand, the company learns even more about human tasks and behavior, something that’s essential to successfully integrating robots into our lives. But the downside of that are the ever-present privacy concerns.
In 2010, Google research scientist James Kuffner came up with the idea of cloud robotics, robots that leverage the Internet, crowdsourcing, and open-sourcing to expand their processing power and knowledge base. It’s not hard to see how a cloud-connected robot that is gathering data in your home from all its various sensors could be a perambulating privacy violation waiting to happen.
“When you have a robot in your house, and it’s taking lots of data, and sharing it to the cloud, that’s a big privacy issue,” said Goldberg. “I’m absolutely not Pollyanna about this, it’s not unvarnished good news.”
Another is the issue of jobs, and what happens to the people working in positions that could be replaced by robots. Many of those jobs, especially in the home care and service industries, are low-paying positions. What kinds of employment can they seek if they’re replaced by an automaton?
“Is Google concerned? I don’t know. Anybody in this business has to be concerned,” said Bajcsy. “My answer is education, education.”
Whatever shape Google’s robotics plans take, and no matter how buffeted the neophyte robotics industry is by the news, Goldberg sounded a cautiously optimistic tone by quoting Sophocles: “Nothing vast enters the life of mortals without a curse.”