AI to the rescue

Artificial Intelligence technologies like Machine Learning and Deep Learning can play a pivotal role in Cyber Security

If Cybercrime were an economy, at USD 6 trillion, it would be the world’s third-biggest after the US and China. It is also growing at a faster rate and is likely to be worth USD 10.5 trillion by 2025. (source: cybersecurity ventures)

According to a report, Ransomware attacks could cost their victims USD 265 billion annually by 2031. The same report predicts that a new episode will likely occur every two seconds as perpetrators progressively refine their activities. (source: cybersecurity ventures)

The growth of technologies like Cloud, Internet of Things (IoT), 5G, and Big Data combined with the move towards remote working – initially propelled by the pandemic but now becoming an essential feature in most organisations has meant that the number of end-user devices, networks, and user interfaces has continued to grow at an exponential rate.

This has created heightened vulnerabilities which attackers and malicious actors have sought to exploit. They have become more aggressive and savvier. What’s more, according to corporate cyber security experts, the number and frequency of machine-speed attacks have increased significantly. These ransomware and other automated attacks propagate and/or mutate very quickly and are virtually impossible to neutralise using human-dependent response mechanisms.

In a survey of 850 corporate cyber security professionals, more than half said that their analysts are getting overwhelmed by the number, pace and frequency of attacks. Moreover, almost a quarter of them said they were not even able to successfully investigate such incidents, opening the doors for more possible breaches. (source: capegemini)

In this environment, corporations have realised that if they have to protect their IT hardware, software, networks and data organisations, they need to turn to Artificial Intelligence (AI), Machine Learning (ML) and Deep Learning (DL) technologies.

The reason is not difficult to understand. Given the explosive traffic growth, cyber analysts are challenged to identify deviations. AI and ML make it easier to analyse these patterns and identify potential anomalies quickly.

Uses of AI in Cyber Security

There are seven areas where large corporations are incorporating AI in the cyber security infrastructure. The highest importance is generally accorded to Network Security, followed by securing data and endpoints. But AI has also found significant usage in security identity and access, apps, besides cloud and IoT infrastructure.

Detection: Currently, AI is more used for cyber threat detection as machine learning, or deep learning-based detection allows organisations to continuously evolve detection parameters, using behavioural analysis to identify anomalies.

Prediction: The use of AI in predicting cyber threats is a rising trend as it can scan through massive amounts of data to make predictions based on how the system has been trained. This can enable preventive actions to preempt attacks.

Response: The use of AI to respond to a breach is still nascent but, according to most experts, is likely to skyrocket in the years to come. AI, for example, can reduce the time taken to develop a virtual patch or create new defensive mechanisms against evolving threats.

Advantages of using AI in Cyber Security

The enhanced use of AI in Cyber Security has three distinct advantages:

• Enhanced Speed: AI can significantly reduce the time taken to respond to detect threats and breaches. When under a cyber-attack – which today most large corporations are almost 24X7 -- every nanosecond counts. AI helps in this critical aspect.
• Reduced Costs: Since AI allows corporations to understand and reuse threat patterns to identify new threats, it reduces the time and effort to identify incidents, investigate them, and remediate threats. Combine this with the enhanced speed of response and reduce the potential damages from a breach.
• Heightened Efficiency: According to most experts, there is a severe scarcity of cyber security talent compared to the size of the threats most large organisations face. The use of AI ensures that much of the grunge work is taken care of, and analysts can concentrate more time analysing the incidents identified by the AI cybersecurity algorithms.