F5 enhances AI data protection with real-time leak detection tools

F5 has unveiled new data leak detection and protection features within its Application Delivery and Security Platform (ADSP) to assist organisations in safeguarding sensitive data when working with artificial intelligence (AI) applications.

The newly introduced functionalities aim to increase visibility into encrypted traffic, enabling businesses to identify and block unauthorised data sharing and defend against shadow AI.

F5's updates include improvements to its AI Gateway and BIG-IP SSL Orchestrator, intending to simplify security management and regulatory compliance across modern hybrid and multicloud environments.

AI security and compliance

The ADSP enhancements are designed to address difficulties encountered when businesses adopt AI solutions and move towards hybrid cloud infrastructures.

Sensitive information is increasingly traversing encrypted channels and exposing organisations to new risks if accessed or transmitted via unapproved AI tools. F5's solution seeks to tackle these risks by offering real-time detection and mitigation of data leakage within complex, encrypted environments.

F5 officials highlighted the scale of the issue, noting the pressures facing business leaders in balancing the adoption of AI with stringent requirements for data protection. Kunal Anand, Chief Innovation Officer at F5, said:

"The core tension in every boardroom today is the race to adopt AI versus the mandate to protect the firm's data. Forcing a choice between the two is a losing strategy. We're eliminating that choice. By providing deep visibility into encrypted AI conversations, we're giving leaders the controls to stop data leakage and govern AI use, effectively turning the CISO from a gatekeeper into the primary enabler of secure innovation."

According to F5, the newly deployed data leakage detection and prevention capabilities for the F5 AI Gateway use technology acquired from LeakSignal to analyse AI prompts and responses.

The system can spot sensitive information - such as personal or confidential data - and act according to customer-specified rules, including redacting, blocking or logging those details.

Key features detailed

The F5 AI Gateway enhancements include real-time detection and policy enforcement to protect and redact sensitive data as it flows into AI environments, and to prevent the unauthorised departure of information from approved environments.

Detailed reporting and audit logs can be integrated into existing Security Information and Event Management (SIEM) tools, giving organisations clearer oversight and simplifying compliance procedures.

Planned for release later in 2025, F5 intends to upgrade its BIG-IP SSL Orchestrator, introducing expanded AI data protection by providing real-time, deep visibility into encrypted traffic. This update will enable organisations to classify, detect and block unauthorised uses of AI and sharing of sensitive data while allowing permitted data traffic through. Reports and dashboards are to be centralised, supporting audit and investigation activities.

The company emphasises that these changes offer the ability for organisations to:

• Detect, classify and stop data leaks in encrypted and AI-driven traffic in real time.
• Prevent risks from unauthorised AI use (shadow AI) and the exposure of sensitive information.
• Apply consistent security and compliance policies across all applications, APIs and AI services.

Addressing regulatory demands

With the push for more effective data controls, organisations are contending with increased scrutiny from regulators and growing obligations to demonstrate robust management of AI infrastructure.

The adoption of these F5 technologies is positioned to help companies maintain compliance across environments that leverage on-premises, cloud and multicloud deployments.

The update to F5's AI Gateway allows organisations to define their own sensitive data policies, leveraging up-to-date inspection technology to intercept and manage data at key points of transit. This is anticipated to reduce risk by ensuring that sensitive details do not inadvertently leave the network or become exposed to third-party AI systems.

F5 confirmed that these new measures will support compliance and audit readiness, meeting evolving expectations as more enterprises adopt AI at scale within critical workflows and regulatory oversight increases in response to rapid AI integration.