Can Artificial Intelligence & Machine Learning Help in Cyber Security?

Digital developments bring with them a set of unique challenges. The new-age digital platforms are not free from vulnerabilities, in fact, they sometimes end up increasing them. The attack surface has significantly increased, giving greater leeway to cyber criminals. Cyber threats continue to grow in frequency and sophistication as the threat landscape quickly evolves. This is a cat and mouse game where we are trying to update our security devices and at the same time some new threats are coming in. In the constantly changing threat landscape, digital transformation technologies such as Artificial Intelligence are being increasingly looked at to mitigate and stave off the threat factor. Artificial Intelligence now has taken the space where no ordinary analytics can deliver results. There are a lot of use cases in AI which help security professionals take the right decisions. The simplest one is early fraud detection.

With the influx of new technologies such as AI and ML, threats could be automated,” says Anshuman Pund, Head Information Security & ISO Compliance, IDBI Intech Limited.

You could also write an AI on user behavioral analytics to identify anomalies of user behavior, which would then be used to identify potential disgruntled employees and thereby preventing insider threats. So, the scope of AI is unlimited and is in a nascent stage and is still developing,” says Amit Pradhan, CTSO , Chief Privacy Officer and SVP - Tech Security, Vodafone India Limited.

In fact, the explosion in data due to digital transformation has also added a new dimension to the threat landscape. Attackers continue to get more sophisticated. Attackers spend significantly to bypass security checks and tools. Security teams need to evolve rapidly to stay ahead of the attackers. Today we are unable to make use of data available through conventional norms. There is a data overload, there is a need to orchestrate all different elements of security to play harmoniously to make the music pleasant to business and leaders. Today, it is not synchronized hence creating a huge gap between business expectation and security solutions,” says Kalpesh Doshi, chief information security Officer, APAC, Group IT, Capgemini Technology Service India Limited.

In fact, modern day attacks are not only capable of crippling the operations of an organization, but can cause irreparable damage to a company’s reputation as well as costing leading to losses running in billions of dollars. There are examples a plenty of the increasing severity of such attacks. Global pharma giant Merc lost $300 million dollars in one quarter only due to a cyber-attack.

With the increase in zero-day attacks that constantly morph and evolve, it is becoming extremely challenging for the existing security solutions to effectively tackle them. There is a distinct need for solutions that keep pace with the attackers. AI has the capability to properly perceive the cyber threat environment to identify events and take appropriate action. It is particularly good at recognizing patterns and anomalies within them, which makes it an excellent tool to detect threats. Machine learning is often used with AI. It has the ability to “learn” on its own, based on human input and results of actions taken.

Together with AI, ML can become a tool to predict outcomes based on past events. With the attacks becoming more and more voluminous and sophisticated, there is a growing need to become proactive and use predictive models to thwart potential threats; which can be facilitated through the use of AI and ML,” says Satish Warrier, CISO, Stock Holding Corporation of India Ltd.

In fact, AI and ML are slowing becoming integral parts of the cyber security set-ups these days. Also, for AI to be successful we will need to standardize processes to gather and analyze data across all devices / applications. Voice and Visual search-based queries will continue to grow and innovation is expected in this space, e.g., Google Home. IT security roles will have to evolve to manage AL and ML technologies.

A version first appeared on DynamicCIO.com