IBM’s Chief Watson Security Architect Jeb Linton demonstrating to University of Maryland, Baltimore County student Lisa Mathews how to teach IBM’s Watson the language of security, (Mitro Hood/Feature Photo Service for IBM)

Watson, IBM's computer brain, has a lot of talents. It mastered "Jeopardy!," it cooks, and even tries to cure cancer. But now, it’s training for a new challenge: Hunting hackers.

On Tuesday, IBM Security announced a new cloud-based version of the cognitive technology, dubbed “Watson for Cybersecurity.” In the fall, IBM will be partnering with eight universities to help get Watson up to speed by flooding it with security reports and data.

The plan as of now is for Watson to process up to 15,000 documents about digital security a month -- including everything from blog posts to videos -- so that it can get a feel for the sometimes esoteric terminology of the cybersecurity world.

Students at the partnered universities will help with that by initially annotating documents so that Watson will be able to interpret the material on its own down the line.

IBM's computer brain Watson has participated on Jeopardy, been programmed to cook, and accomplished things many believed it couldn't. Take a look at what IBM has in store for Watson's future. (IBM Watson)

The end goal is a big data approach to cybersecurity that will have Watson automatically scour vast troves of security research at a rate human operators couldn't possibly manage to investigate when something fishy hits a victim’s computer systems.

"It's automating the hunt,” explained Caleb Barlow, vice president for IBM Security.

Right now, security practitioners are overwhelmed with a flood of alerts about possible threats to their networks, according to Barlow. “They can't sift through all the data that's coming at them -- many of which are false positives,” he said.

According to one 2015 report from the Ponemon Institute, more than half of the time security staff spends investigating malware alerts is wasted on inaccurate intelligence or false alarms -- costing organizations $25,000 per week on average.

That's where Watson can help. Watson won't necessarily replace those staffers, according to Barlow. Instead, the system could help them prioritize the almost never-ending flow of alerts heading their way -- hopefully shutting down potential attacks faster and putting others on the alert.

"When we can stop an attack early on and then tell everyone else, we don't only block it -- we're also playing offense by stopping the bad guys from attacking anyone else," said Barlow. 

Barlow also said that the company's relationships with universities could help train up more human talent to fend off attackers.

The University of Maryland Baltimore County, for instance, is creating a new Accelerated Cognitive Cybersecurity Laboratory via a partnership with IBM this fall. The lab will focus on using machine learning and systems like Watson to solve cybersecurity problems. UMBC is also one of the universities that will help train Watson this fall, along with the Massachusetts Institute of Technology, California State Polytechnic University Pomona, Pennsylvania State University, New York University, the University of New Brunswick, the University of Ottawa and the University of Waterloo.