Varonis unveils AI Shield to defend sensitive data in real time

Varonis has introduced AI Shield, a product designed to provide persistent AI risk defence for data security by monitoring and remediating data exposure in real time.

AI Shield operates by continuously analysing an organisation's AI security environment, observing interactions between AI systems and data, and dynamically regulating data permissions to prevent the exposure of sensitive information resulting from inadequate data security practices.

The solution leverages Varonis' patented permissions analysis algorithms, which take into account factors such as data sensitivity, data staleness, user profiles, and more, to determine contextually which data should be restricted from AI access. This approach is intended to protect organisations even when they have not properly configured access controls.

David Bass, Executive Vice President of Engineering and Chief Technology Officer at Varonis, highlighted the complexity introduced by the prevalence of AI in data security. He said, "AI makes the data security challenge much more urgent and complex. AI Shield gives our customers the confidence to deploy AI with both preventative and detective controls that require zero setup and maintenance. It's always on, always learning, and always working for you behind the scenes to prevent breaches and compliance violations."

AI Shield provides several features aimed at ensuring secure AI usage, including real-time risk analysis to pinpoint the sensitive data exposed to AI, automated risk remediation that continually addresses data exposure at scale, and behaviour-based threat detection to identify abnormal or malicious actions. The system also offers round-the-clock alert response to investigate, contain, and stop data threats.

According to the company, organisations with poor data security posture face the risk of employees or AI systems potentially accessing large volumes of data that were not intended for their access. Varonis states that its AI Shield is designed to address this issue by offering constant monitoring and intervention capabilities.

The company emphasises that protecting data access by AI tools is now a central component of broader data security measures. The product aims to help employees safely interact with AI tools, ensuring that only authorised users and AI agents have access, while all activity is monitored and any suspicious behaviour is detected and flagged.

Varonis positions AI security as inseparable from data security, asserting that solutions like AI Shield are required to address the increasing complexity of protecting sensitive data in environments where AI-driven processes are regularly utilised.