IBM's Watson computer, which beat champions of the quiz show Jeopardy! a year ago, will soon be advising Wall St on risks, portfolios and clients.
Citigroup, the third-largest US lender, is Watson's first financial services client, IBM said yesterday. It will help analyse customer needs and process financial, economic and client data to advance and personalise digital banking.
IBM expects to generate billions in new revenue by 2015 by putting Watson to work.
The technology giant has already sold Watson to health-care clients, helping WellPoint and Seton Health Family analyse data to improve care.
IBM executives say Watson's skills - understanding and processing natural language, consulting vast volumes of unstructured information, and accurately answering questions with humanlike cognition - are also well suited to the finance industry.
Financial services is the "next big one for us," said Manoj Saxena, the man responsible for finding Watson work.
IBM is confident that with a little training, the quiz-show star that can read and understand 200 million pages in three seconds can make money for IBM by helping financial firms identify risks, rewards and customer wants mere human experts may overlook.
Banks spent about US$400 billion ($489 billion) on information technology last year, said Michael Versace, head of risk research at International Data Corp's Financial Insights, which has done research for IBM.
Watson the financial assistant will be delivered as a cloud-based service and earn a percentage of the additional revenue and cost savings it is able to help financial institutions realise.
Watson, including its work in the health-care and finance industries, will contribute "a portion" of IBM's target of US$16 billion of analytics revenues in 2015, Saxena said, and that portion will "have a B next to it".
Watson may add US$2.65 billion in revenue in 2015, Ed Maguire, an analyst at CLSA in New York, estimated in a research note.
IBM, the world's biggest computer-services provider, reported revenue of US$107 billion in 2011 and earnings of US$13.06 a share. The company ended 2011 with US$11.9 billion in cash.
IBM shares closed above US$200 for the first time yesterday, factoring in stock splits.
Watson "can give an edge" in finance, said Stephen Baker, author of books The Numerati and Final Jeopardy, a Watson biography.
"It can go through newspaper articles, documents, SEC filings, and try to make some sense out of them, put them into a context banks are interested in, like risk."
In addition to Citigroup, Armonk, New York-based IBM has been working with financial institutions teaching Watson the language of Wall St, and adding content including regulatory announcements, news and social media feeds. "It's not selling them software, it's selling them outcomes," Saxena said.
Watson offers a "more global" picture by looking beyond financial data, Saxena said. For example, Watson can comb prospectuses, loan performances and earnings quality while also uncovering sentiment and news not in the usual metrics before offering portfolio recommendations. It can also monitor trading, news sources and Facebook to help a treasurer manage foreign exchange risk.
Some of the biggest financial institutions have already built big data centres. IBM is competing with most other major technology companies to sell them tools to analyse and use accumulated information, Versace said. "Apparently Citi gets it - analytics is the new core in competitive banking," Versace said.
Watson gives IBM "a huge marketing edge" in the race among tech giants including Google and Microsoft to obtain intelligence for businesses by teaching machines to understand sentences and paragraphs rather than searching for single words or phrases, said author Baker.
Parts of Watson could already be combined with other IBM technologies to help banks with regulatory compliance by surveying internal documents and flagging those that seem amiss, Baker said. Watson was designed to be loaded with information rather than grapple with a live streaming feed, making it likely IBM will partner it with other technologies.
Watson may have applications for telecommunications companies, and perhaps even call centres, said Saxena, who joined IBM when it acquired his company Webify Solutions in 2006.
IBM plans to use Watson in financial services "mostly for portfolio risk management, they're not going to do stock picking", CLSA's Maguire said.
Still, Watson isn't perfect. It is weak in languages other than English, and its processing of social media streams from platforms including Facebook and Twitter can be sluggish. The lag is "getting shorter", Saxena said.
A year ago last month, about 15 million viewers watched Watson beat former Jeopardy! champions Ken Jennings and Brad Rutter - a highly publicised victory in artificial intelligence that IBM always aimed to apply to the business world.
Watson had to learn how people speak and write, and evaluate its level of understanding, said Eric Brown, a member of the team that built him in an IBM research facility in New York. Precise answers delivered with confidence distinguish Watson from web searches, Brown said.
"I'm sure they use all this stuff internally to manage their own portfolio," said CLSA's Maguire. "IBM's treasury is bigger than a lot of trading desks. If they went into asset management they would kick ass, but that's not what they do."
* IBM's Watson computer can read and understand 200 million pages in three seconds.
* Last year Watson beat former champions Ken Jennings and Brad Rutter on US TV quiz show Jeopardy!.
* Now IBM is applying the technology to the finance and healthcare industries.
* Watson is expected to contribute billions to IBM's bottom line by 2015.