White-collar robots roll into schools, hospitals and offices

A cross between a Dalek from the long-running BBC science fiction series Doctor Who and R2-D2 from Star Wars could be the key to preventing the deaths of children in school shootings across the US, as robots move out of the factories and into service industries.

Shiny and white, the autonomous guard robot patrols an area, using data from optical and sound sensors along with licence plate recognition to feed information to law enforcement authorities or private security guards.

Stacy Dean Stephens and his team at Knightscope came up with the idea after the Sandy Hook shootings at an elementary school in Connecticut in 2012. As a police officer in Texas, Mr Stephens thought law enforcement officers needed better technology.

"After Sandy Hook, all the analysis showed that if you were able to get officers inside the school to the subject just one minute earlier, as many as 12 lives could be saved," he says. "I wanted to look at how we would go about using technology to help achieve that goal."

The Knightscope K5 Autonomous Data Machine uses a combination of robotics and predictive analytics to determine when the police or security company should be alerted to a threat.

Five feet tall and packed with sensors, it can carry more than a police officer can fit on his or her back belt and save them from the most "monotonous, boring and mundane work", Mr Stephens says.

The robot is being tried out on the campuses of Silicon Valley tech companies, which are more able to experiment than schools. Knightscope has received interest not just from the education sector, but also from organisers of the upcoming Olympics in Brazil and Tokyo, large US security companies and other businesses from the Middle East to China.

The start-up, still only a dozen people, is at the forefront of a new era in robotics, as smarter technology begins to creep into more white collar and even professional industries. Mr Stephens says autonomous technology is on the "verge of a revolution, a tremendous change in the coming decade", comparing it with the computer revolution of the 1980s.

Professor Erik Brynjolfsson, who along with MIT Sloan School of Management colleague Andrew McAfee wrote the recent book The Second Machine Age, says technology is being developed that could disrupt hundreds of millions of jobs.

He says that while the first machine age, the industrial revolution, replaced people's muscles with machines, the second is replacing cognitive tasks completed by humans.

"It is going to have some similarities to the first machine age - there will be tremendous bounty - but there will also be very important differences," says Prof Brynjolfsson. "When you replace muscle work with machines, you still need humans to make decisions about what is to be done, making human work more valuable, but in the next wave it is not as clear whether machines complement or substitute humans."

Machines can now diagnose breast cancer better than humans, first-year lawyers have been almost replaced by legal discovery software and self-driving cars have a mental picture of the blind spots of every vehicle on the road and can spot one slowing from further away.

Robots are often cheaper than humans, work longer hours and can be made to do the less safe tasks. Aethon, a Pittsburgh-based company, sells its Tug robot - a self-driving trolley - to hospitals, which use it to transport medicines, samples for the lab, meals, laundry and waste.

Aldo Zini, chief executive of Aethon, says hospitals obtain "nice hard dollar savings" with a 150 per cent return on investment over a few years. But he adds: "The biggest advantage is that robots are good for doing things that are dangerous or not very suitable for a human to do - wheeling a 500lb laundry cart, picking up infectious waste, transporting very expensive chemotherapy drugs."

Aethon's revenues are almost doubling year-on-year, as it expands to 140 US hospitals and begins to win contracts internationally, including in Denmark, Germany and Canada. It has been contacted by companies from hotels to shops asking if the Tug could work for them.

But some robots come in less robot-like shapes. Blue Prism, a UK-based company that works for the back office of customers including Barclays and the Co-operative Bank, sells a robot that looks just like automated software. What makes it a robot is that it fills in forms and uses computer systems just like a human, without any extra changes to the IT platform.

Alastair Bathgate, chief executive, says robots can be trained to do the work of tens of thousands of back office employees. Clients can still have a human answering the phone, but when they take down the details of someone's lost credit card, for example, they can pass the form-filling to a robot.

"This is not about a load of P45s [the form given to UK workers when they lose their jobs]; it is about reallocating costs," he says.

Robots, he adds, are not very good at apologising to customers, but are more accurate than humans at completing the paperwork. "They cannot replicate empathy, human intelligence, sympathy, creativity or entrepreneurialism, but they can carry out humdrum activities."

For Illah Nourbakhsh, a professor at the Robotics Institute of Carnegie Mellon University in Pittsburgh, one of the most important things companies must think about when introducing robots is the effect on existing employees.

"When you add robots you don't take away all the people. Psychologically, it can have an impact on people's relationship with robots and management," he says. "It introduces a whole new extinction dynamic."

-------------------------------------------

Robots keep down costs, speed up results

From Brazil to Bulgaria, Canada to China, more than 30,000 employees work for Sutherland Global Services, a business process outsourcer. Now, they have been joined by an invisible robot.

The outsourcing industry has focused on cutting costs by moving routine work offshore to markets with cheaper labour.

But Ian Barkin, global head of innovation at the company's Sutherland Innovation Labs, says that, whereas it can save between 20 and 40 per cent by shifting work to a developing economy, it can reduce costs by up to 70 per cent if it uses robots that do not need to be paid at all.

He says: "In the past couple of years, automation has come to the fore. Rather than figuring out ways to enable people to do things more efficiently, the industry is looking at automating people out of the process."

Sutherland Global Services, based in Rochester, New York, uses a "robot" - in the form of automation software - from Blue Prism, a UK-based company, to complete tasks involving high volumes of structured data, such as accounts payable and receivable, order management and some HR functions.

Mr Barkin says that, as well as the cost savings, using a robot speeds up auditing and gives companies a better understanding of what tasks take the most time. And, if disaster strikes, computers elsewhere can be brought into play, rather than having a whole office out of action.

But he warns that while Blue Prism's robots can be instructed to use computer systems quickly and easily, companies need to make sure they are trained very carefully. "If there's some shoddy training for employees it is a bit of a mishap, but if a miscoded robot is scaled across many robots, it is very efficient at creating havoc."

Charles Sutherland, executive vice-president, research at outsourcing research company Horses for Sources, says robots can help companies that have been doing "totally preposterous" things, such as having two people enter the same data to ensure it is accurate.

However, he says much outsourced work requires juggling exceptions to company rules, which indoctrinated robots are not good at.

"You can't entirely replace outsourcing, because robots still can't operate with a degree of judgment to manage exceptions," says Mr Sutherland.

He also says that estimates of the potential savings for companies using robots should be taken with a "large pinch of salt", as they often do not take into account the cost of integrating computer systems and a number of variables between different tasks.

© The Financial Times Limited 2014. All rights reserved.
FT and Financial Times are trademarks of the Financial Times Ltd.
Not to be redistributed, copied or modified in any way.
Euro2day.gr is solely responsible for providing this translation and the Financial Times Limited does not accept any liability for the accuracy or quality of the translation