Report: GenAI Flaws Could Negatively Impact Organisation’s Cybersecurity Strategies

A recent report by Sophos, a leader in cybersecurity solutions, highlights both the potential and the risks of Generative Artificial Intelligence (GenAI) in cybersecurity. The survey, which involved 400 IT leaders, revealed that while 65% of organizations have adopted GenAI, 89% of IT leaders are concerned about flaws in these AI tools that could jeopardize their cybersecurity strategies. This uncertainty raises alarms about the vulnerability of organizations relying on GenAI for protection.

The Sophos X-Ops team also released findings indicating a shift in the way cybercriminals are utilizing AI. While many still exhibit skepticism toward GenAI, there has been a growing trend of cybercriminals leveraging AI for automating routine tasks, such as crafting bulk emails and analyzing data. Some are even using it to enhance spam and social engineering attacks, suggesting that GenAI’s impact on cybersecurity is multifaceted and not limited to legitimate uses.

Chester Wisniewski, Director of Global Field CTO at Sophos, emphasized that while GenAI tools offer significant potential to improve security processes, they still require human oversight for proper context and comprehension. He advised IT leaders to approach GenAI with caution, adopting a “trust but verify” stance to mitigate the risk of misuse or unintended consequences.

The report also sheds light on concerns about over-reliance on AI in the cybersecurity field. Many IT leaders are worried about the loss of accountability, with 87% fearing that AI could diminish human oversight in cybersecurity decisions. Additionally, there is widespread concern about the pressure to reduce cybersecurity teams due to unrealistic expectations surrounding AI’s capabilities to replace human operators.

Finally, the report addressed the complexities surrounding the cost of GenAI in cybersecurity. Seventy-five percent of IT leaders agree that the costs associated with GenAI tools are difficult to quantify, making it challenging for organizations to evaluate their return on investment. Despite this, organizations of all sizes are continuing to integrate AI into their cybersecurity infrastructure, recognizing its potential to improve both efficiency and security.

Source: This day