With the recent major ransomware attacks in the news, it's apparent that companies need to implement better defences, most likely in the form of artificial intelligence. 

While old techniques for identifying malware involve matching the code of malicious viruses against a database of known malware, newer techniques characterise the malware by its behaviour. More specifically, looking for combinations of behaviours and observable characteristics usually associated with malicious intent - for example, by quarantining a program disguised with a PDF icon to hide its true nature.

There are still issues with this approach, however, and that's when we turn to machine learning, which is a form of artificial intelligence. A security system analyses samples of good and bad software and determines what combination of factors is likely to point to malware.

Essentially, society is currently using anti-virus software that still relies on malware databases that aren't necessarily kept up to date. The future of anti-virus relies heavily upon discarding these databases and instead opting for machine learning.