UK Edition
IBM to use Jeopardy! winning machine Watson to fight cyber attacks
Perhaps better known for pitting its wits against the winners of a US television quiz show, IBM’s artificial intelligence asset is now being unleashed against cyber attackers.
Watson, originally designed as a question answering machine, beat two of the most successful Jeopardy! contestants in 2011 and has since mastered tasks including providing music recommendations, helping with water conservation in California and personalising cancer care. It has now made another addition to its impressive skillset as the computing giant has launched a special version for helping to fight cyber attacks.
The new look Watson can analyse the latest research into potential threats and apply its understanding to suspicious activity within companies’ computer systems.
IBM has said the system can help thwart major hacks which have become a growing concern, hitting companies including Yahoo, Lloyds and TalkTalk. Artificial intelligence can also fill a critical skills gap in the industry, which is expected to have a global shortfall of 1.5m security professionals by 2019.
The company says the system can help thwart the major hacks that have become a growing concern, hitting companies including Yahoo, Lloyds and TalkTalk. AI can also help fill a critical skills gap in the industry, which is expected to have a global shortfall of 1.5m security professionals by 2019.
The Watson security machine can solve problems in minutes that would take a human weeks, and save up to 20,000 hours a year spent chasing false alarms, IBM said. Having been trained for a year, it is now ready to be used within companies.
"It can help companies find an advantage over the growing legions of cyber criminals and next generation threats," said Dennis Kennelly, vice president of technology at IBM Watson. "Our investment in Watson for cyber security has given birth to several innovations in just under a year.
"It combines the unique abilities of man and machine intelligence will be critical to the next stage in the fight against advanced cybercrime."
Watson is an AI machine that can understand human language. The cyber security version has been trained on over a million security documents and can now parse swathes of regularly updated research and apply it to its protection of a company's network.
IBM isn't the first security company to employ AI in the fight against online crime. Darktrace, a British cybersecurity company valued at $400m (£320m), uses machine learning to understand the nuances of companies' computer systems and fight cyber attacks as they happen.
The use of AI in cyber security is expected to triple in the next two years. Just 7pc of professionals work with it today, according to IBM research.
"Today's sophisticated cybersecurity threats attack on multiple fronts to conceal their activities, and our security analysts face the difficult task of pinpointing these attacks amongst a massive sea of security-related data," said Sean Valcamp, chief information security officer at Avnet, one of the companies working with Watson.
"Watson makes concealment efforts more difficult by quickly analysing multiple streams of data and comparing it with the latest security attack intelligence to provide a more complete picture of the threat."
Jon Oltsik, senior principal analyst at the Enterprise Strategy Group, said: "It is well timed as organizations have continuing difficulty hiring experienced security analysts, while the threat landscape is increasing significantly."