IBM stated that Rensselaer’s faculty background in Web science, big data and artificial intelligence “uniquely situates” the university to best use the Watson system. RPI faculty, graduate students and undergrads are all expected to work with the system for research and innovation, particularly in finance, IT and business analytics, as well as preparatory work for “future high-impact, high-value careers in analytics, cognitive computing, and related fields.” As the first non-IBM team working on the Watson system, there’s an expectation that efforts at the university could hone the Watson’s systems reasoning and cognitive capabilities as well.
The 15-terabyte version of Watson at RPI has more storage capacity than its game show winning predecessor, and enables 20 workers to access it at once. IBM and Rensselaer have had a long history of collaboration, especially at the university’s Computational Center for Nanotechnology Innovations lab, where the new Watson system will be housed.
Universities nationwide are expanding and deepening traditional analytic degree programs to address the perceived shortage in data management jobs, and Rensselaer President Shirley Ann Jackson said in a statement that she expects “Watson will give our students an advantage as they compete for the best jobs in Big Data, analytics, and cognitive computing.”
Wednesday’s announcement is the first introduction of Watson as an entire system available outside of IBM’s walls. Big Blue continues to tinker with a commercial version of the system. Much of the vendor's hardware and software for the system is already generally available, such as its Power 750 servers, though a combined, commercial Watson system could be ready "in a few years," according to an IBM spokesperson.
In its public close-up in February 2011, the supercomputer decimated its human competition on TV quiz show “Jeopardy!” in 2011. Since then, IBM has applied analytic capabilities behind Watson in health care and business number crunching, as well as other displays of its accurate data association capabilities in prop game shows against college students.