AI and Data Protection in 2019

    Jan 15, 2019 11:11:00 AM Keeran Networks Data Security, Cybersecurity

    The amount of personal and business data being uploaded into the digital universe continues to expand at a sta­ggering rate; a recent study by IDC estimated that the total amount of digital data will reach 40 ZB by 2020.

    What is a zettabyte? If a kilobyte (kB) is nominally 1000 bytes, a zettabyte is 1 000 000 000 000 000 000 000 bytes. Or 10007 bytes. Or 1021 bytes. You get the idea. Either way you look at it, that’s a lot of cat pictures on Instagram, and a lot of critical data that will needs protection.

    With new privacy regulations, increasingly complex cyberattacks, and the adoption of cloud-based disaster recover strategies, we can expect to see some big changes in how we approach data protection in 2019.

    A big part of this will come down to the implementation of AI – it will increasingly allow organizations to move past real-time insights and start predicting and averting unplanned downtime, proactively back up critical data to the cloud, and provide further insight into what data is truly essential based on the frequency of employee access.

    The AI Arms Race between Cyberattacks and Cyberdefence

    Like any tool, AI can be used for good or bad. And those who are responsible or corporate security – including CIOs, CISOs and CROs – must be aware of how AI is changing the nature of their jobs. Advanced machine learning, deep learning, and neural networks are what enable computers to discover and analyze patterns – and in some cases, this capability is used to find and exploit vulnerabilities.

    Concerningly, AI is being used to increase the sophistication and effectiveness of cyberattacks. Researchers at IBM discovered invasive and targeted cyberattacks powered by artificial intelligence that used geolocation and facial recognition as triggers.

    Malware and Ransomware is starting to learn as it spreads, AI can be used to coordinate large scale global attacks, and data analytics are being employed to tailor phishing attacks based on a target’s browsing activity. The established model of creating a ‘perimeter’ security network for your organization is becoming increasingly ineffective as attacks focus on new ways to target individuals and gain access to data.

    But it’s not all bad news – the same abilities that are making cyberattacks more potent is making cybersecurity more effective.

    Related Blog: Does your Law Firm have a Data Breach Response Procedure in Place?

    Augmenting IT Security with AI – Three levels of implementation.

    AI augment existing security measures by enhancing detection and response capabilities, reducing complex and time-consuming manual inspections and interventions, and redirecting human efforts to focus more on proactive problem-solving and supervisory tasks.

    Here are three levels where AI could enhance an organization’s existing IT security systems and practices:

    Prevention and Protection:

    For a long time, security flaws and vulnerabilities were only revealed during an attack or through time-consuming testing. However, AI shows great potential in being able to stop cyberintruders though the application of advanced machine learning to discover existing security flaws and deploy appropriate solutions in real time, and allow human input to adjust and refine the methodology.


    This is a big one. Leveraging the ability to analyze extremely large amounts of data, AI algorithms will be able to quickly notice & detect abnormalities. This means that these systems will have an ‘understanding’ of what normal network and system activity should look like, and without being told, accurately identify suspicious activity.

    As the internet of things grows, it will be become increasingly important that physical equipment and devices are safeguarded against hijacking. One challenge facing the adoption of this detection will be dealing with the laws and regulations governing data use.


    In addition to automating manual tasks such as reviewing log files, AI can be used to implement intelligent responses to attacks based on shared knowledge and machine learning.

    Even today, AI based security has the ability to employ “traps” that duplicate the target environment, fooling the attacker while providing a safe environment in which to attempt to identify the source. These response systems can quickly create barriers to isolate data within the networks and redirect attackers away from valuable data.

    Is it time for an IT Security Audit?

    If all this talk of cybersecurity has you double-guessing the safety of your organization’s data and IT infrastructure, we can help you out.

    Keeran Networks offers comprehensive Security Audit Services that cover both risk and compliance, identifying vulnerabilities for you to act on, and ensuring your data handling practices meet industry regulations.

     Get in Touch


    Keeran Networks

    Written by Keeran Networks