NSS Labs Names Keysight Lead Partner in New AI Protection Systems Security Testing Initiative

Independent methodology in development for validating AI security guardrails against thousands of adversarial attacks.

AUSTIN, Texas, March 23, 2026 /PRNewswire/ -- NSS Labs today announced that Keysight Technologies has joined its new AI Protection Systems (AIPS) security testing initiative as lead partner, supporting the development of one of the industry's first independent evaluation programs dedicated to testing AI security guardrail technologies.

As artificial intelligence becomes foundational to digital transformation across industries—including finance, healthcare, government, and critical infrastructure—the security and integrity of AI systems has emerged as a global priority. Organizations are rapidly deploying AI models and applications, yet the technologies designed to secure and govern their use—often referred to as AI guardrails, AI firewalls, or AI runtime protection systems—have not yet been independently validated through standardized testing. 

To address this gap, NSS Labs is launching a comprehensive independent evaluation program dedicated specifically to AI Protection Systems (AIPS)—security platforms designed to enforce policy, prevent misuse, and defend AI models and applications from adversarial attacks. The initiative aims to establish a transparent, technically rigorous methodology that benchmarks how effectively these systems protect AI deployments against real-world threats while maintaining policy enforcement and operational integrity. 

Keysight is a foundational partner, supporting the development and execution of this groundbreaking AI security validation program.

The NSS Labs AIPS methodology evaluates products across seven dimensions of AI security, including malicious input and prompt attacks, output risks and sensitive data exposure, system resilience under adversarial conditions, policy enforcement accuracy, agentic AI and tool invocation security, observability and audit capabilities, and performance and scalability impact. 

Across these layers, the methodology includes hundreds of thousands of individual test case executions designed to bypass, manipulate, exploit, or overwhelm AI Protection Systems. Each scenario is executed using multiple attack samples and variations—including prompt injection attempts, jailbreak techniques, obfuscated prompts, sensitive data extraction attempts, exploit generation requests, RAG poisoning attacks, API privilege escalation attempts, and agent tool misuse scenarios. 

"AI is rapidly becoming core infrastructure for the digital economy, and with that comes an urgent need for independent validation of the technologies designed to protect it," said Vikram Phatak, CEO of NSS Labs. "With Keysight joining us as lead partner, we are bringing together our deep expertise in testing along with Keysight's global innovation solutions that will help the industry understand how well AI protection systems actually perform against real-world threats."

"AI is quickly becoming foundational infrastructure, and trust in these systems must be earned through transparent, independent validation," said Ram Periakaruppan, Vice President and General Manager, Network Test & Security at Keysight. "Keysight's strength in building scalable, real-world test environments and generating actionable performance insights positions us to help shape how AI security is measured. We're proud to partner with NSS Labs to advance a more resilient and trustworthy AI ecosystem."

By combining adversarial testing, policy validation, system robustness analysis, and operational visibility checks into a single structured framework, the NSS Labs AI Protection Systems test aims to establish a credible benchmark for independent AI security validation and provide enterprises with objective data on the effectiveness of technologies designed to secure AI.

Feedback for the methodology is currently being accepted from enterprises and security vendors. Please reach out to [email protected] for a draft copy if you would like to provide comments. The AIPS methodology will be published in April.

Executives from NSS Labs are attending the RSA Conference. Please contact us if you would like to schedule a meeting.

Additional Resources:

(White Paper) AI Security Beyond the Model: What Enterprises Need to Care About – and Why
(White Paper) Evaluating Enterprise AI Security: Questions Every Buyer Should Be Able to Answer

About NSS Labs

NSS Labs delivers research-backed insights through its advanced testing platforms, empowering enterprises, service providers and security vendors to make informed, evidence-based cybersecurity decisions. By handling the heavy lifting of testing for effectiveness, performance, and suitability, NSS Labs helps clients move beyond assumptions to gain actionable clarity. Its auditing and governance services offer continuous assurance that deployed security technologies are performing as expected, protecting investments and supporting accountability. For more information visit nsslabs.com    

SOURCE NSS Labs