After winning Jeopardy and getting a gig in the health care industry, IBM’s Watson has been “hired” by Citigroup to see how Watson’s deep content analysis and ability to learn could apply in the financial sector. According to the IBM press release, Citi wants to use Watson “to help advance customer interactions and improve and simplify the banking experience.” And, presumably, also boost Citi’s profits.

The idea is that Watson will use Citi data to analyze customer needs and identify both risk and reward factors for the bank. The relationship between Citi and IBM is public knowledge, although it’s probably not the only “employment” that Watson has in the financial world. Its searching ability would certainly benefit both commercial and investment banks, which are so heavily reliant upon huge amounts of unstructured data.

Every time I read about Watson’s technology and its application in industries as disparate as health care and finance, I wonder how it could be utilized within libraries. Could Watson supplant the discovery services? Could it really tell whether the person asking a question was an undergraduate, a tenured professor, a bench chemist, a corporate librarian, a financial analyst, or something else? Could Watson revolutionize how librarians do collection development? Could it be the next generation of text mining?

More to the point, could companies in the information industry afford Watson? I’m pretty sure that libraries can’t.