Artificial intelligence (AI) – a computer science term that describes machines and computers which mimic such cognitive human functions as learning and problem solving – and machine learning (ML) – the study of algorithms and statistical models that computers use to perform tasks without being given explicit instructions – are undisputedly two of the most transformative technologies to have been developed in recent years. Both have already impacted a whole range of sectors with the scope of their application getting wider and wider every year.
Costs getting higher
While both AI and ML have already become mainstream technologies in many fields, their use in fighting cybercrime has until now been mainly limited to only some of the largest organizations. However, according to experts, this will now be changing fast. Indeed, we can safely assume that in the coming year’s AI- and ML-based cybersecurity tools will become indispensable almost everywhere, argued Lovlesh Chhabra, Vice President, Membership Platforms, at Verizon Media.
The growing costs related to security breaches are one of the main reasons for this. Much as many companies may still be reluctant to invest more in security tools, they may want to consider the damage a failure to secure vital assets entails, added Chhabra who in the years 2011-2017 worked as Director of Product Management – Consumer Platforms at Yahoo where he was responsible for developing the company’s Account Key solution. The solution helped protect Yahoo’s users during a historic breach into Yahoo email accounts.
A 2019 report conducted by the Ponemon Institute and sponsored by IBM Security found out that the average total cost of a data breach amounts to over USD 3.9 million. In the US, it stands at almost USD 8.2 million. For its part, a report released by Juniper Research last year predicts that the global business losses resulting from cybercrime data breaches will increase by nearly 70% over the 2019-2024 period to reach more than USD 5 trillion. Chhabra pointed out that the costs are not only about clean-up expenses but also include brand and reputation damage.
AI battling AI
Importantly, the risks related to data breaches are magnified by the fact that malware keeps getting more and more destructive and that cybercrime is today often sponsored by states. What is more, cybercriminals themselves are increasingly often using AI in their attacks with a view to defeating authentication measures and avoiding being detected. All of this means that cybersecurity AI needs to be employed to fend off such advanced attacks. Indeed, studies show that if properly used the latest technologies can significantly lower cybercrime risks and costs.
How can they help?
First of all, ML can sift through huge amounts of data to identify assets that need to be protected, as well as to find errors in code and other vulnerabilities, Chhabra claimed. He added that AI-based tools can also be useful for fixing bugs. Crucially, AI can be instrumental in enhancing already existing security practices such as authentication methods.