It is official. The robot apocalypse will come with the howling fury of an angry customer service call. Designers are currently working over the next six months on feeding data which is comprised of collection calls showing human beings at their worst. There’s a special kind of zen required of those who work in customer service. Now, a company in New Zealand wants to design a computer program that can mimic the hatred of angry callers in order to help those same customer representatives deal with riled up customers.
The project is named Radiant, after a supercomputer in Isaac Asimov’s Foundation series that could predict the future. While the real-life Radiant won’t be quite so omniscient, its designers at the technology firm Touchpoint hope it will be able to accurately simulate millions of angry customers to help companies figure out what makes people fly off the handle, writes Michael Bingemann for The Australian
New Zealand-based technology firm Touchpoint Group is developing the world’s angriest artificial intelligence machine that it hopes will one day help big banks, telcos and insurance companies defuse explosive episodes in customer service.
The new machine learning research project, which Touchpoint is investing $500,000 to develop, is being built with input from one of Australia’s big four banks, which is supplying reams of real-life customer interactions that have been collated over the past two years. Telecommunications companies and insurance firms are also contributing data.
Data scientists in Australia and New Zealand will spend the next six months uploading the dataset into the platform and tweaking its learning algorithms with an expectation that it will be live by the end of the year.Once complete, the project will simulate hundreds of millions of angry customer interactions that will help companies better understand the behaviours and processes that trigger customer outbursts.Touchpoint Group chief executive Frank van der Velden said the research would help with the complex task of understanding how customers were affected by the various products, systems, policies, processes and people they interacted with in the lead-up to reaching breaking point.Mr van der Velden said the program would constantly run “what if” scenarios to see if a particular scenario was likely to enrage or benefit the customer.
“The end goal is to build an engine that can recommend solutions to companies — and we’re talking about the people at the frontline here — how they can improve particular issues that customers are facing,” Mr van der Velden said.
“This will be possible by enabling our AI engine to learn right across a whole range of interactions of what has and has not worked in past examples.”
In the Foundation series, Prime Radiant was a supercomputer that could predict the future behaviour and development of humanity through the analysis of history, sociology and mathematical statistics. Mr van der Velden said Touchpoint’s program would be attractive to any company that had to deal with customer service complaints.
“Companies don’t have the numbers of staff to go through this manually. It’s very difficult. Take a bank for example, they receive a hell of a lot of data every day. But it gets to a point where that dataset grows so large that it becomes meaningless unless you can interpret it. That’s where Radiant will fit in,” he said.
“We’re not in the business of managing complaints; we are in the business of managing issues that might turn into complaints. We’re at the top of the cliff, not at the bottom. This will allow companies to better predict and identify those issues.”
Radiant works by examining data from the worst of the worst customer service calls and determining what factors and experiences could set someone off in any given scenario. Touchpoint is working with one of Australia’s biggest banks and several insurance companies and telecommunications firms that are supplying the customer service data that is embittering Radiant towards everyday life. By the time the program is up and running, Radiant will be able to react angrily and irrationally to telemarketers and customer service representatives-in-training who will have to try and calm the computer down. They hope to complete the program by the end of the year.
From the description of the project, it sounds like Radiant has been tasked with identifying patterns of what sets off different emotions. According to Stefan Weitz, a senior director of search at Microsoft who knows quite a bit about machine learning, the future of machine learning will be in teaching robots to identify patterns — critically analyzing queries rather than pinging the web to find results.
The designers are currently working on feeding Radiant data over the next six months, which is comprised of collection calls showing human beings at their worst. Radiant will be digesting the data, and sifting through the calls to determine what variables could set people off at any given point on a customer service call. The computer will then turn what it has learned on mankind in order to help train bank telemarketers and customer service representatives in conflict management. The developers are hoping to complete Radiant by the end of the year. The wrathful robot comes at a high price though. But if it works, it might make your next angry phone call to a company go just a little bit smoother.