This page was machine-translated and may differ from the original. View original
Development of AI Safety Case Establishment Methodology Compliant with ISO/PAS 8800
Keysight Technologies announced on the 22nd that it has signed a joint research agreement with the Centre for Assuring Autonomy (CfAA) under the University of York in the United Kingdom to support the safe deployment of AI within SDVs.
This agreement focuses on generating evidence data to demonstrate the safety and reliability of AI systems, and narrowing the gap between AI safety research and practical engineering applications.
Development of ISO/PAS 8800 Compliant Methodology
The core of the joint research is summarized in three aspects.
△Development of evidence-based approaches supporting compliance with ISO/PAS 8800 requirements △Exploration of measurable safety assessment methodologies based on academic research and industry standards △Establishment of practical frameworks for generating auditable AI safety evidence.
The derived methodologies are expected to be utilized by engineering teams to efficiently demonstrate AI safety throughout the entire product lifecycle.
Simon Burton, Chair Professor of Systems Safety at the University of York, stated, "The automotive industry is at a critical turning point where AI is becoming central to vehicle functionality, and it is essential to evaluate these systems with a robust data-driven approach."
He added, "CfAA has been disclosing frameworks and guidelines already being utilized by industrial safety experts in the transportation sector, and this collaboration will serve as an opportunity to translate AI safety principles into engineering practice."
Reflection in AX1000A Product Development
Lucas Klotze, Head of Keysight's Automotive AI Solutions Center, stated, "Automotive companies need practical means to build confidence in AI-based systems. By combining the expertise of the University of York with Keysight's AI validation framework, we will support the deployment of AI technologies that comply with international standards such as ISO/PAS 8800."
The research results will be reflected in the development of Keysight's AI Software Integrity Builder (AX1000A Keysight AI Software Integrity Builder), with plans to strengthen capabilities supporting AI safety argumentation establishment and validation activities.
To request a correction, reply or follow-up report on this article, see how to file a request. Previously published statements are collected in corrections & replies.













