Singapore (VNA) - Singapore has proposed a new international standard to harmonise testing methods for generative artificial intelligence systems, aiming to strengthen the foundation for trustworthy AI evaluation.
The proposed standard, ISO/IEC 42119-8, is the first international standard of its kind. It was announced as Singapore hosts a global AI standardisation plenary meeting this week.
The proposed standard aims to ensure trustworthy AI by standardising benchmarking and red teaming methodologies.
The standard builds on Singapore’s earlier initiatives in AI governance and testing, including the IMDA’s AI Verify toolkit, the Starter Kit for testing large language model-based applications, and the Global AI Assurance Sandbox. It also aligns with broader national efforts such as EnterpriseSG’s ISO/IEC 42001 accreditation programme and contributions to international documentation on real-world AI use cases.
The proposal will be discussed at the 17th ISO/IEC JTC 1/SC 42 plenary meeting held in Singapore from April 20 to 24.
Co-organised by the IMDA (Infocomm Media Development Authority) and Enterprise Singapore (EnterpriseSG), the meeting brings together more than 35 national bodies and over 250 AI experts from markets including the US, the UK, China, Japan, Germany, France and the Republic of Korea.
Alongside the plenary, the IMDA and EnterpriseSG are also running a series of capacity-building programmes, including workshops on AI standards capability, governance applications and AI assurance practices. These sessions bring together policymakers, standards bodies and industry stakeholders to explore how AI frameworks can be translated into real-world implementation./.
