Freely subscribe to our NEWSLETTER

Whilst Generative AI has undoubtedly been utilised by cybercriminals in their assaults on organisations, it is ultimately a key tool in cyber defence as well. Gen AI’s ability to learn, replicate and predict data patterns, whilst also recognising consistent signs of cyber threats or vulnerabilities, is why cybersecurity is currently one of the top use cases for AI, as corroborated in IDC’s whitepaper, "The Business Opportunity of AI".

Brian Sibley, Solutions Architect at Espria highlights that the most successful modern threat intelligence strategies have, for some time, incorporated reliable, secure Gen AI tools to anticipate and mitigate growing threats against businesses.

"Generative AI is a powerful tool that cybersecurity professionals are utilising to develop new security tools to detect and prevent cyberattacks. For instance, the technology has been employed to recreate realistic phishing emails, mimicking cybercriminals in order to train and upskill employees; behavioural data is then harvested and utilised to proactively monitor and protect against attacks.

“The real-time threat detection capabilities also enable rapid response, minimising the impact of cyberattacks and reducing the potential damage caused by latency issues in traditional cybersecurity measures. Despite the misconception that GenAI is a replacement for human input, instead, it aims to enhance capabilities and unlock greater potential. Implementing these Gen AI powered tools has enabled cyber security professionals to execute their roles more effectively by automating mundane tasks and offer better support to the humans hunting for the more crucial threats.

However, Sibley believes it is critical to understand that not all GenAI tools are set up to protect businesses from data leaks. “Unprotected, public tools, which use publicly available data as part of their learning model, such as OpenAI (ChatGPT) and Llama, might appear useful due to their ease of access, but in terms of security pose a potential risk for data leaks.

"Without key access controls, an unregulated GenAI tool can give the wrong employees access to business-critical information, increasing the potential for in-house data leaks and security risks. Similarly, there are also other darker variants of AI tools which hackers can access and fine-tune for malicious purposes. These models behave like an unrestricted version of the tool that allows malware creation and advanced phishing attacks.

“Security teams can utilise these tools to great effect, but this does not necessarily translate directly into employee usage. Proper data leak mitigation procedures need to be enforced more than ever so that business data is completely secure from potential threats, both inside and outside the organisation. Thankfully, our experts, in conjunction with our cybersecurity partners are ready and available to help, so business owners can ensure their employees can only access data appropriate to their role and whichever GenAI tool the business chooses.

"Partnerships between Microsoft and Veeam and leading security specialist providers such as, Mimecast and Sophos, have meant that shared telemetry leverages greater predictive powers within the AI used in their products. These shared resources offer businesses far greater opportunity to protect their data assets and the use of GenAI within these tools will only enhance this advantage. GenAI is set to transform businesses in 2024, so ensuring that any AI tool used within the business is secure in of itself should be a priority.”