European IT Leaders Divided on the Value of Generative AI in Cybersecurity, Study Finds
A new study by Corelight reveals a lack of consensus among European IT leaders regarding the value of generative AI (GenAI) in cybersecurity. The network detection and response (NDR) specialist surveyed 300 IT decision-makers (ITDMs) in the UK, France, and Germany to compile the report titled “Generative AI in Security: Empowering or Divisive?”
Mixed Sentiments on GenAI in Cybersecurity
The findings indicate that GenAI is viewed with both optimism and concern. Nearly half of the respondents (46%) are proactively exploring ways to integrate the technology into their cybersecurity strategies. Conversely, a similar percentage (44%) expressed concerns about data exposure and enterprise silos, making the use of GenAI in cybersecurity challenging or unfeasible. Additionally, 37% of respondents believe that GenAI is “not safe to use in cybersecurity.”
Security Concerns and Risks
Security risks associated with GenAI are well-documented. For instance, Samsung banned the use of ChatGPT in 2023 after employees shared private meeting notes and source code with the tool. Sensitive customer data or intellectual property shared with such tools could potentially be accessed by other users, posing significant security, privacy, and compliance risks.
However, Corelight asserts that these challenges can be addressed if GenAI tools and the underlying large language models (LLMs) are designed with privacy in mind. The company’s commercial NDR solution claims to establish a “functional firewall” between customer data and GenAI features.
“GenAI’s adoption is hindered by concerns over data confidentiality and model accuracy. As models improve in reasoning capacity and cybersecurity knowledge, and as more LLM deployments include structural privacy protections, GenAI is set to become integral to security operations,” said Ignacio Arnaldo, director of data science at Corelight.
Potential Benefits of GenAI in Cybersecurity
Half (50%) of the ITDMs surveyed believe GenAI’s most significant impact on cybersecurity could be providing alert context and analysis for security operations (SecOps) teams. Other potential uses cited by respondents include:
- Maintaining compliance policies (41%)
- Recommending best practices for domain-specific languages like identity and access management policy (36%)
- Managing unstructured vulnerability information (35%)
- Providing remediation guidance (35%)
- Handling unstructured network connection and process information (32%)
Despite the mixed sentiments, the study suggests that with proper implementation and privacy safeguards, GenAI could play a crucial role in enhancing cybersecurity operations.
Source: infosecurity-magazine.com
