Cognitive Computing Points To Smarter Future

IBM's Watson computer system, powered by IBM POWER7, competes against Jeopardy!’s two most successful and celebrated contestants -- Ken Jennings and Brad Rutter -- in a practice match held during a press conference at IBM's Watson Research Center in Yorktown Heights, NY on January 13, 2011. Watson will compete against Jennings and Rutter in the first-ever man vs. machine Jeopardy! competition, which will air on February 14, 15 and 16, 2011, with two matches being played over three consecutive days.

Perhaps when future historians look back to our time, they will point to the 2011 television show of the popular U.S. TV quiz Jeopardy as the turning point in the dominance of the human species.

After a well-publicized three-night marathon, IBM’s Watson supercomputer defeated two champion, but human, players. It was a spectacular display of what IBM calls “cognitive computing.”

Now that same technology is coming to a bank near you. In fact, it is already here.

Jeopardy! practice game. Watson beat two human champions to be crowned the winner

Watson is being used to manage risk and provide personalized guidance and investment options at financial institutions. ANZ Global Wealth in Australia is using Watson to observe the types of questions coming from both customers and financial advisors. DBS Bank in Singapore is applying Watson to its wealth management business to improve the advice and experience delivered to its customers. The Inland Revenue Authority of Singapore is using Watson to help answer users’ questions about taxation. And Spain’s CaixaBank is working with IBM to help teach Watson Spanish.

Finance is one of the areas that lends itself to cognitive computing, says Sridhar Iyengar, Distinguished Engineer at IBM Research and part of the Watson team, and a guest speaker at Sibos 2015.

“The real area for Watson and cognitive computing is to help accelerate and scale human expertise by including unstructured data into decision-making.” It is estimated that unstructured data — the contents of a report, a video clip, perhaps a tweet — accounts for more than 80% of the world’s data.

Watson analyzes high volumes of data and processes information more like a human than a computer — it understands natural language and can generate hypotheses. Those strengths can be put to work in banks; but where will that leave employees?

Sensitive to the worries, Iyengar is keen to stress that Watson takes very much a supporting, not a leading, role. “We see Watson more as a cognitive assistant, working with the human, to make the human more effective.”

Aim Is To Make Employees More Effective

“Imagine an expert financial advisor. If you take a look at a typical financial services firm they may have literally thousands. A very small percentage of them are really, really good — perhaps maybe the top five or 10%.

“They are there because of the years of experience, the knowledge they have accumulated, and the insights they have gained into decision making.

“Our goal with Watson is really to bring that capability to the other 80%-90% of the people who may not have had the benefit of the years of experience. We will find a way to scale their knowledge with this augmented intelligence so that they can serve the customers better.”

It is a view echoed by Pedro Gaspar, Head of Live Banking for Novabase, a Portuguese software development house. Its Wizzio financial advisor support tool is powered by Watson. Given the impact that the financial crisis has had on employment, is Wizzio just another way for banks to lay off staff?

Gaspar denies that is the motivation. “We were very concerned to make an inclusive solution. The aim is to enhance the work of a financial advisor — it is not about replacing him or her, it’s about bringing more value. We are helping people become more productive, and so more valuable to the bank.”

Financial Wizzard

When Portuguese software development house Novabase was looking for a new tool, named Wizzio, to help financial advisors support their clients they knew they needed some kind of intelligent agent — but which one?

“It wasn’t just a question-and-answer interface,” says Pedro Gaspar, head of Live Banking for Novabase, “We wanted something that would evolve and would learn from experience. That took out of the equation solutions like Microsoft’s Cortana and Apple’s Siri. They are very smart, they understand natural language, but they are just assistants.”

“You can tell Watson that a fund is an investment which is true — but it’s not a full truth. After a while Watson will understand that it is not just an investment, it is more than that.”

What Watson lets advisors do, says Gaspar, is to explore options. It digests and assesses information, and produces weighted recommendations. “Watson places out the options and then the advisor, based on his knowledge chooses the best option.”

Gaspar was quick to refute any suggestion that Wizzio is a way for banks to slash headcount. “Wizzio does two things. It brings more quality to the advice itself. That adds a lot of value.

“[And] Advisors are expensive. That means the bank can only use them for very wealthy clients. If you can lower the advising costs, that mean you can reach more people.”

Comments

comments

Authors

Related posts

Top