The growing hype around machine learning technology seems to be creeping into every aspect of enterprise IT and cyber security is no different.
Defined as a way to provide computers with the ability to learn without the need for programming, the term Machine Learning seems to be this year’s big buzzword. In fact today you’d be hard-pressed to find a security vendor that doesn’t claim to incorporate some form of machine learning or AI technology into their threat prevention solution. So how much is true and how much is smoke and mirrors?
Despite being a concept first envisioned back in the 1950s by Alan Turing, the lack of computing power has held research teams from making any significant jump towards realising the technology. Nowadays we all rely on machine learning without even knowing it – Google’s search engine is powered by machine learning APIs, as is Facebook’s personalised news feed.
So now that it’s here, how can it be put to work to enhance cyber security defences and what do IT professionals need to look out for? “The important thing to remember with machine learning from a security perspective is that you don’t know what bad looks like before bad happens.” HPE’s Chief Technologist of Hybrid IT Security Simon Leech told us.
He suggests machine learning can help security experts make huge leaps in malware detection and biometrics as well as keeping out hackers. “A good user behaviour analytics system can start to correlate different aspects of a log-on process together. So this can be simple things like where is the user logging on from? What is the browser type? What is his physical location?”
So how do security professionals navigate this new realm of solutions and will they really make their business any more secure?
Listen to the full podcast below: