Heaptalk, Jakarta — The Singaporean Government unveiled a generative AI testing toolkit called AI Verify Project Moonshot through the Ministry for Communications and Information on Asia Tech x Singapore (ATxSG) 2024 (05/31). This launch aims to address security and safety challenges associated with large language models (LLMs).
Singapore’s Minister for Communications and Information, Josephine Teo, stated that Moonshot is one of the world’s first open-sourced tools to bring red-teaming, benchmarking, and baseline testing together in an easy-to-use platform, a testament to the country’s commitment to harnessing the power of the global open-source community in addressing AI risks.
Project Moonshot is a user-friendly toolkit that provides intuitive results about the quality and safety of a model or application. This open beta toolkit, developed through a collaboration with several prominent tech firms, including DataRobot, IBM, Singtel, and Temasek, is designed to be easily understood by non-technical users.
This project is also part of a vital move towards global testing standards. AI testing organizations, such as the AI Verify Foundation and MLCommons, are joining forces to build a common safety benchmark suite. “We are proud that leading AI testing organizations AI Verify Foundation and MLCommons are coming together to support common testing benchmarks for AI safety, building on innovative efforts like Project Moonshot,” said Peter Mattson, MLCommons President and co-chair AI Safety working group.
AI Verify Foundation and ML Commons signed a memorandum of intent (MOI) on 29th May to further the collaboration. The MOI intends to positively impact AI safety by providing model and application developers with a globally accepted approach to safety testing for generative AI.
