Independent methodology in development for validating AI security guardrails against thousands of adversarial attacks.
NSS Labs has announced that Keysight Technologies has joined its AI Protection Systems (AIPS) security testing initiative as the lead partner, marking a major step toward establishing independent validation standards for AI security technologies. The collaboration reflects growing urgency across industries to secure AI systems as they become foundational to digital transformation.
As organizations increasingly deploy AI across sectors such as healthcare, finance, government, and critical infrastructure, concerns around AI security, governance, and data protection are intensifying. Technologies designed to safeguard AI systems – commonly referred to as AI guardrails, AI firewalls, or runtime protection platforms – have rapidly emerged, but until now, they have lacked standardized, independent evaluation frameworks.
To address this gap, NSS Labs is launching a comprehensive testing program dedicated to AI Protection Systems. The initiative aims to deliver a transparent, technically rigorous benchmarking methodology that evaluates how effectively these platforms defend AI models and applications from real-world cyber threats while maintaining operational integrity and policy compliance.
At the core of this program is an advanced testing framework that assesses AI security across seven critical dimensions. These include protection against malicious inputs and prompt injection attacks, prevention of sensitive data exposure, resilience under adversarial conditions, accuracy of policy enforcement, security of agentic AI and tool interactions, observability and audit capabilities, and overall system performance and scalability.
The testing methodology is designed to simulate real-world threat environments at scale. It incorporates hundreds of thousands of attack scenarios, including prompt manipulation, jailbreak techniques, obfuscated queries, data extraction attempts, exploit generation, retrieval-augmented generation (RAG) poisoning, API privilege escalation, and misuse of AI-driven tools. This large-scale simulation approach enables enterprises to understand how AI security systems perform under complex and evolving threat conditions.
Vikram Phatak, CEO of NSS Labs, emphasized the importance of independent validation in the evolving AI landscape. He noted that as AI becomes critical infrastructure, organizations need reliable benchmarks to assess the effectiveness of security technologies designed to protect it. The partnership with Keysight brings together deep testing expertise and advanced innovation capabilities to address this need.
Ram Periakaruppan, Vice President and General Manager of Network Test & Security at Keysight, highlighted the role of scalable testing environments in building trust in AI systems. He stated that transparent, real-world validation is essential for ensuring that AI deployments remain secure, reliable, and aligned with enterprise requirements.
By combining adversarial testing, policy validation, system robustness analysis, and operational visibility into a unified framework, the NSS Labs AIPS initiative aims to establish one of the first credible industry benchmarks for AI security validation. This effort is expected to help enterprises make informed decisions, strengthen AI risk management strategies, and accelerate the adoption of secure, trustworthy AI technologies.
Recommended Cyber News :
- Yubico and Delinea Address Agentic AI Accountability Gap
- Netzilo Launches AI Edge for Enterprise AI Security
- Microsoft Unveils Agentic AI Security Strategy
To participate in our interviews, please write to our CyberTech Media Room at info@intentamplify.com
