Let’s see, we have a company that already knows everything about us, has possibly the world’s largest computer network, has recently built one of the biggest artificial-intelligence teams in the world–a company so powerful that it feels the need to soften its dominance with the informal motto, “Don’t be evil.”
And now Google Google–yes, of course we’re talking about Google–has bought a military robot company call Boston Dynamics. Not just any robot maker this time–after all, it has already quietly bought seven others over the past year, apparently to provide former Android chief Andy Rubin another chance at a moonshot project. No, unlike the other robot makers, this company makes machines by the names of BigDog, Atlas, and Cheetah that can variously outrun Usain Bolt and hurl cinderblocks 17 feet.
So, we’ve got the potential for killer robots that know where you live and can outrun you when they find you. What’s not to like?
All jokes about Skynet, Terminators, and Robocops aside, the latest acquisition raises a serious question about what Google has in mind. It looks for all the world like it’s pursuing yet another seemingly crazy side project that has nothing to do with its mission to organize the world’s information and make it universally accessible and useful. It’s now trying out self-driving cars, home package delivery, wearable computers, and anti-aging technologies.
Clearly it’s time for Google to update its mission statement, not to mention the “Ten things we know to be true,” a list that includes such outdated gems as “It’s best to do one thing really, really well.” More recently, CEO Larry Page has made the case that Google’s broader mission, which he noted was mentioned in Google’s first letter to shareholders, is to improve people’s lives. A little more specifically, the New York Times has noted, Page has “argued that technology should be deployed wherever possible to free humans from drudgery and repetitive tasks.” So maybe there’s a method to this particular moonshot that isn’t completely disconnected from the reality of Google’s current $40 billion advertising business.
Consider self-driving cars, which looked rather off-mission themselves. And while a couple of years in now, they still don’t seem to fit squarely into Google’s core mission or business, we can now see their usefulness in Googley concerns such as collecting data about the real world, producing new insights into artificial intelligence that might be more broadly applicable in search, and even saving people time driving (and avoiding getting killed in accidents) so they can spend more time in front of a screen.
OK, that last one is a stretch, and the others may not be all that convincing either, as one commenter below argues. But it’s still worth asking how robots might fit Google’s raison d’etre. Here’s how Rubin and others inside Google, according to the Times, view the near-term possibilities:
The company’s expected targets are in manufacturing — like electronics assembly, which is now largely manual — and competing with companies like Amazon in retailing, according to several people with specific knowledge of the project. A realistic case, according to several specialists, would be automating portions of an existing supply chain that stretches from a factory floor to the companies that ship and deliver goods to a consumer’s doorstep.
OK, but that still sounds a little far afield of the core business. It could be a sign that Google is indeed serious about its delivery service, or even that it might consider the technology as a way to get even more active in making its own computer servers.
But take a look at the other part of Boston Dynamics’ business:
The company also develops tools for human simulation. DI-Guy is a human simulation product used for simulation-based training, UAV training, law-enforcement training, mission-planning and many other applications. DI-Guy has become the defacto standard for human simulation as used in military applications; it is used by leading organizations world-wide, including all branches of the US Armed Forces. Digital Biomechanics is a physics-based human simulation product used to design and evaluate equipment used on the human body, such as backpacks, helmets, body armor and the like.
Those tools suggest more human-centered applications that could improve the way people interface with computers, wearables, and information generally. Clearly they’re military-oriented now, but it’s easy to imagine broader applications that relate to self-driving cars, Google Glass, and even information retrieval and analysis.
For now, this is all sheer speculation. The fact is that Google, like AT&T, IBM, and Xerox in previous decades, has monopoly-like profits that it can use to do research into areas that go far beyond its current business. Whether robots or self-driving cars or wearable computers become significant businesses for Google is less important than the fact that today, it’s willing to spend the big bucks to push forward in these seemingly unrelated areas. One of them may well turn out to evolve into something that helps Google’s current business or redefines it in a coherent, profitable way. But we can’t know yet.
Meantime, take a look at some videos of these robots and see if you can figure out why Google’s so interested in them. Here’s Cheetah running really fast (don’t try this on your home treadmill):
Here’s BigDog tossing a cinderblock (incoming!):
This is Atlas, humanoid enough to give you Terminator nightmares:
And not to go out on too creepy a note, here’s a “weaponized” BigDog (we can only hope they don’t show up in Pamplona):