As artificial intelligence changes the financial services industry, should advisors feel threatened? No, computer experts say. They see the current wave of technology as an opportunity to provide better service.
Much has been made of the research that finds automation will steal work from millions of Canadians, including a University of Toronto study from last year.
But it's hard even for experts in the field of artificial intelligence (AI) to see wealth managers entirely replaced by chatbots, or apps for financial planning and investment.
Still those involved in the rapidly changing discipline of computer science say the financial industry is fertile ground for increasingly intelligent automation.
"Definitely all the big banks are very excited about AI," says Stephen Piron, co-founder of DeepLearni.ng, a Toronto-based tech company that builds artificial intelligence software for clients including financial institutions.
The financial industry has long been moving toward increased automation, from ATMs to online brokerages. Moreover, the investment world has leveraged big data and artificial intelligence for the past decade.
"On trading desks, this sort of thing is a little bit old hat, but that's kind of a crazy, esoteric world," Mr. Piron says. "Now it's moving down to the more retail banking stuff and it is all about helping businesses understand their customers better."
Already these advances are moving into the wealth management industry with robo-advisers, which employ AI techniques to create automated portfolios based on the characteristics of investors.
Yet the industry is just getting started, says the founder of a Vancouver-based company that provides robo-advisory services as well as wealth management AI software for independent wealth management firms.
"Fundamentally, we view wealth management as a software problem," says Davyde Wachell, CEO of Responsive AI.
By "software" Mr. Wachell isn't referring just to an operating system that is installed on a desktop or the cloud.
"It's about organizing principles, problem-solving capabilities and managing behaviour."
Right now he argues the industry's software model is antiquated and does not serve consumers well.
"It's a distribution system and the advisor's key role is to on-board the clients and get them into" an investment to generate fees, he says.
Rather than giving individual consumers the best strategy for their needs, the current model is geared toward generating as much profit as possible, he says. In turn, this has helped to fuel the rise of low-cost alternatives such as robo-advisers as investors find they are better served by skipping the middleman in portfolio management: the human advisor.
Responsive AI is upping the ante when it comes to wealth management robotics.
"The next stage is financial planning that is driven by context and having goal-based systems," Mr. Wachell says. "This is where the computer understands your situation and starts to give you concrete things you can do on the short-term and long-term horizon."
For example, wealth management software is being developed to bring to bear all the data available about an investor – Facebook posts, spending habits, investment choices, pins on Pinterest, tweets, etc. – using AI to find patterns to customize a financial plan and investment strategy.
The benefits of developing enhanced software for the industry go beyond helping to understand clients better. Responsive AI, for example, is currently providing software that handles the back-end office and compliance work for a wealth management operation.
"We're automating a lot of processes, getting rid of the paper cuts for advisors and making it easy to keep on side with compliance," he says. "All the things that drain the blood out of advisors and their ability to focus on clients."
But the future involves creating AI that draws on a discipline of social psychology called psychometrics that uses statistics to build profiles of investors.
"These personality traits can be mapped back into the planning strategies that the advisor is going to make for the client," he says.
Mr. Wachell says the financial industry has been rather slow in adopting these advances compared with large tech companies, such as Google and Facebook, and marketing agencies. That's a consequence of the banks' conservative nature. These publicly traded firms tend to focus more on the next earnings season – driven by the existing software mode –rather than the long-term picture, he says.
But the future will increasingly involve automated services using AI models such as machine learning where computers can make inferences about client behaviour and needs based on large amounts of data.
That's not say Canada's largest financial institutions don't have AI on their radar. The Royal Bank of Canada, for example, has invested tens of millions of dollars in developing AI, including cutting-edge versions of it like deep learning, pioneered in Canada by University of Toronto professor Geoffrey Hinton.
"Where the industry is today, we're just scratching the surface," says Gabriel Woo, vice-president of innovation at RBC.
Already AI helps analysts sift through massive amounts of data – news stories, financial reports and economic outlooks – to help assess securities and markets.
But for the time being, AI remains rather dumb when it comes to the human touches of wealth management.
"AI can help you process and gather a lot more data, but how do you balance the conflicting information that will come through in the data?" he asks. "That still requires judgment, and that's still something that AI is not very good at."
Where Mr. Woo and others see a lot of promise on the wealth management side is using advances in computer science to augment the abilities of advisors to understand clients and provide better portfolio construction beyond fitting investors into an aggressive, conservative or balanced box.
It's not unlike a concept used by the U.S. military – one of the leaders in AI – which involves merging the technology with humans, sometimes referred to as centaurs. In wealth management, the idea is that flesh-and-blood advisors would be supported by AI technologies.
"AI would help people become more effective at what they do, but it wouldn't actually on its own do things it's not instructed to do," Mr. Woo says, adding this process will accelerate in the next few years.
Central to the advance of AI will be specific processes like machine learning, in which a computer learns and in turn becomes smarter over time. Even more intriguing is the potential of its subset, a discipline called deep learning, which attempts to emulate the brain's neural networks.
In contrast to current technology that becomes obsolete after a few years, software using these programming techniques could become increasingly intelligent, and in turn valuable, over time as it gains more understanding, Mr. Piron says.
"You can get a computer to truly understand what people want and need at scale – a personalized level."
He likens it to advances in health care with personalized medicine where drugs are developed to match the genetics and other biological characteristics of a patient for more effective treatment.
"Well this would be similar – only it's personalized banking."