Real-world robots still pretty stupid, but…

Opinion

Hey there, time traveller!
This article was published 31/03/2014 (4212 days ago), so information in it may no longer be current.

Robots came into the world as a literary device whereby the writers and filmmakers of the early 20th century could explore their hopes and fears about technology, as the era of the automobile, telephone and airplane picked up its reckless Jazz Age speed. From Fritz Lang’s Metropolis (1927) and Isaac Asimov’s I, Robot (1950) to The Terminator (1984) and WALL-E (2008), and in countless iterations in between, they have succeeded admirably in their task.

Since moving from the page and screen to real life, robots have been a mild disappointment. They do some things humans cannot do themselves, such as exploring Mars, and a host of things that people do not much want to do, such as dealing with unexploded bombs or vacuuming floors — there already are around 10 million robot vacuum cleaners wandering the carpets of the world. They are useful in some areas of manufacturing.

Reliable robots, however, and especially ones required to work beyond the safety cages of a factory floor, have proven hard to make, and robots are still pretty stupid. So, although they fascinate people, they have not yet made much of a mark on the world. That seems about to change. The exponential growth in the power of silicon chips, digital sensors and high-bandwidth communications improves robots the same way it improves all sorts of other products. Three other factors are at work, however.

One is that robotics research and development is getting easier. New shared standards make good ideas easily portable from one robot platform to another, and accumulated know-how means building such platforms is getting much cheaper. A robot such as Rethink Robotics’ Baxter, with two arms and a remarkably easy, intuitive programming interface, barely would have been conceivable 10 years ago. Now you can buy one for $25,000.

A second factor is investment. The biggest robot news of 2013 was that Google bought eight promising robot startups. Rich and well-led — by Andy Rubin, who masterminded the Android operating system — and with access to world-beating expertise in cloud computing and artificial intelligence, both highly relevant, Google’s robot program promises the possibility of something spectacular, though no one outside the company knows what that might be.

Amazon, too, is betting on robots, both to automate its warehouses and, more speculatively, to make deliveries by drone. In South Korea and elsewhere, companies are moving robot technology to new areas of manufacturing and eyeing services. Venture capitalists see a much better chance of a profitable exit from a robotics startup than they used to.

The third factor is imagination. In the past few years, clever companies have seen ways to make robots work as grips and gaffers on film sets — Gravity (2013) could not have been shot without robots moving the cameras and lights — and as panel installers at solar-power plants. More people will grasp how a robotic attribute such as high precision, fast reactions or independent locomotion can be integrated into a profitable business, and eventually some of them will build mass markets. Aerial robots, a.k.a. drones, may be in the vanguard here. They will let farmers tend their crops in new ways, give citizens, journalists and broadcasters new perspectives on events big and small, monitor traffic and fires, look for infrastructure in need of repair and much more besides.

As consumers and citizens, people will benefit greatly from the rise of the robots. Whether they will benefit as workers is less clear, because the robots’ growing competence may make some human labour redundant.

Aetheon’s Tugs, for instance, which take hospital carts where they are needed, are ready to take over much of the work porters do today. Kiva’s warehouse robots make it possible for Amazon to send out more parcels with fewer workers. Driverless cars could displace the millions of people employed behind the wheel today. In the same way that employment in agriculture, which provided almost all the jobs in the pre-modern era, now accounts for only two per cent of rich-world employment, so jobs in today’s manufacturing and services industries may be forced to retreat before the march of the robots.

Whether humanity will find new ways of using its labour, or if instead the future will be given over to forced leisure, is a matter of much worried debate among economists. Either way, robots probably will get the credit or blame.

Robotic prowess to some extent will be taken for granted. It will be in the nature of cars to drive themselves, of floors to be clean and of supplies to move around hospitals and offices. The robotic underpinning of such things will be invisible. Robots will not merely animate the inanimate environment, however. They will inhabit it alongside their masters, fulfilling all sorts of needs. Some, like Baxter, will help make and move things, some will provide care, some merely comfort or companionship. A Japanese robot resembling a baby seal, which responds amiably to stroking and can distinguish voices, seems to help elderly patients with dementia.

The more visible robots are, the better they can help humanity discuss questions such as those first posed in fiction. Is it necessary that wars always be fought by people who can feel pity and offer clemency, and yet who also can be cruel beyond all tactical requirements? Does it matter if the last kindnesses a person feels are from a machine rather than another person? What dignifies human endeavour if the labour of most, or all, humans becomes obsolete?

People, companies and governments find it hard to discuss the ultimate goals of technological change in the abstract. The great insight of Asimov & Co. was it is easier to ask such questions when the technology is personified, when you can look it in the face. Like spacefarers gazing back at the home planet, robots can serve not only as workers and partners, but also as purveyors of new perspectives — not least when the people looking at them see the robots looking back, as one day they will, with something approaching understanding.