Financial companies such as Bank of New York Mellon, BBVA and American Express have become early adopters of a new generation of artificial intelligence and robotic-process-automation technology that automate human tasks.
Their efforts resemble the higher-profile work of Google, Ford and others with self-driving cars and BAE Systems with drones.
Specifically, the banks are creating tech to do chores previously performed by people in operations, wealth management, algorithmic trading, risk management and other areas.
Though it holds great promise in improving efficiency and cutting costs, many worry that low-level jobs will be irretrievably lost in the process. McKinsey predicted in 2013 that AI and robotic banking will displace 110 million full-time workers around the world by 2025.
"I don't think people realize that so many of our jobs will be automated and a lot of manual labor and not terribly intellectual jobs will be replaced," said Christine Duhaime, a Toronto attorney with a practice in anti-money-laundering, counterterrorist financing and foreign-asset recovery and founder of the Digital Finance Institute. "When the average person starts to realize that bank teller jobs are gone, it's going to freak people out."
Robin Hanson, an associate professor at George Mason University, also sees AI and bots replacing jobs but believes society will adjust.
"At some point, computers can do enough to displace an entire profession like travel agents or bank tellers, then a whole bunch of people lose their jobs, [and] then there's a lot more discussion of it," Hanson said. "This is usually part of a long-term trend of more and more lowly jobs getting displaced, but they're small enough that people who are displaced are a small fraction of the whole marketplace; they have some pain and they find new jobs and go on."
In some industries the transition is well underway. The Washington Post began using language-generation technology to produce stories automatically during the 2016 Olympic Games, and it plans to continue to use the technology to cover the presidential campaign.
Jeremy Gilbert, director of strategic initiatives at the Post, has said the paper does not want to replace human reporters with automatons. "We want to free up reporters and editors to add analysis, color from the scene and real insight to stories in ways only they can," he said in a press release announcing the initiative.
Many definitions exist for artificial intelligence. One working description is this: teaching machines to learn and to interact in order to undertake cognitive tasks that were usually performed by humans. This can encompass many technologies that attempt to do things people can do, including machine learning, natural language processing and robotic process automation.
Robotic process automation lets employees configure computer software "bots" to interact with applications and perform high-volume, repetitive tasks. For instance, AI-powered bots in some banks are gathering and analyzing customer information for wealth managers to help them craft investment advice for clients.
Bankers who have been experimenting with and deploying these technologies say they do not want to kill jobs, either.
"I don't think a big goal would ever be to replace people," said Marco Bressan, chief data scientist at BBVA. "In our case, what we want as an institution is to make better decisions, to help our customers make better financial decisions, and if we can do that with the help of technology, then we will."
What will happen is what always happens, he said: people will have to adapt and get retrained to work with the new technologies.
"It's not only that we build the technologies, the technologies rebuild us and the way we redefine the scope of our work," Bressan said.
Further, Bressan believes employees could benefit from this revolution. For instance, work days could get shorter, with AI and bots eliminating time-consuming chores.
Doug Shulman, senior executive vice president and global head of client service delivery at BNY Mellon, sees his bank's work with robotic process automation as "smart automation, using technology to automate a lot more of what we're doing, with a goal of having a better client experience, simplifying and automating what we do, and making it a better place to work so our talent is working on the higher-value-added work, reducing risk and reducing costs."
The bank is using bots for things like account reconciliation.
"Our people are super excited about this," Shulman said. "Nobody comes to work and says, 'I want to pull down a spreadsheet and look at somebody's account and compare it to the account we have on our books and check a box.' People might be willing to do that to make some money, but that's usually not their career aspirations. What's exciting for people is getting to interface with new technology and be part of changing the bank."
Some of the people doing more "mind-numbing" work will shift to client-facing positions, Shulman said. "Our goal is to get as much of the boring work automated as possible so our people can do more interesting, value-added work for clients."
David Weiss, senior analyst at Aite Group, endorses the notion of "peak human" in which bots and AI are used to augment human labor rather than replace it.
Say an insurance company was swamped with claims after a natural disaster. "If I've got 20 people who do claims processing and we're overwhelmed, what do I do?" Weiss said. "Do I add more people? Do I offshore? How does this affect my hiring of employees and location of employees and their skill-sets? Do I have one tier of people who are the experts that the machines learn from and a second tier of people who can follow along and guide and monitor a robotic process automation that's been derived from the experts? How do they do that? They're trying to figure this out. Not a lot of people realize this."
But Weiss also sees companies using AI and bots or "inorganic intelligence" to reduce jobs.
"The cheap, easy path is human replacement, and clearly jobs are lost," he said. "People are instinctively afraid of drones and killer robots. In financial services, their livelihoods could be in danger."
Reimagining the Workforce
Duhaime likes the idea that low-level workers' time could be freed up to do more interesting work than data searches. But she also envisions economic realities taking their toll.
"The bank may end up with the same number of employees, but they'll have to get rid of lower-level jobs and hire people in tech centers to do coding and artificial intelligence," she said. "There's no way you can take all your tellers and make them techies."
Some percentage could be retrained to code, but it's not realistic to think all could, she said.
"We're facing massive shutdowns of branches and infrastructure which will save banks a lot of money, but they are going to have to invest in computer coding, which is a really expensive piece, and that's the reality," Duhaime said. "We'll have a lot more machinery and a lot less people."
Hanson also has trouble believing that banks could keep their current staffing levels as they invest in AI and robotics.
"Of course the long-term standard strategy is to look for the most easy-to-automate jobs, and the easiest-to-automate jobs have always been the ones that are very routine, simple, even mind-numbingly boring; those are all great candidates for things that might be better to automate," he said.
Duhaime sees this trend leading not only to a loss of banking jobs, but taking an especially heavy toll on women in banking, a problem Melinda Gates has called the "sea of dudes." As entry-level jobs are erased in favor of work for programmers and database administrators, women lose.
"The reason there's a sea of dudes is because guys get their computer science degrees and women don't as often," Duhaime said. "We don't have enough women in the space, and there aren't enough teenagers thinking about it who are girls."
While it will take many years for this to change, women should be encouraged to take roles in this field, she said.
"They can be part of the discussion, part of the driving force behind it, as opposed to what I'm really afraid is going to happen, which is they'll have no seat at the table in artificial intelligence for a whole lot of years," Duhaime said. "I don't think we want a society like that, but I think we'll get one if we don't change something quickly."
Editor at Large Penny Crosman welcomes feedback on her posts at firstname.lastname@example.org.