IMDA launches Project Moonshot for AI model safety

DataRobot, IBM, Singtel, and Temasek helped develop the toolkit.

Minister for Communications and Information Josephine Teo has launched AI Verify-Project Moonshot, a testing toolkit designed to address security and safety challenges often associated with the use of large language models. 

AI Verify-Project Moonshot is one of the first open-source tools to bring red-teaming, benchmarking, and baseline testing together in a platform.

“An open beta, Project Moonshot aims to provide intuitive results of the quality and safety of a model or application in an easily understood manner, even for a non-technical user,” authorities said.

IMDA developed the project with DataRobot, IBM, Singtel, and Temasek to ensure that the tool is useful and aligned with industry needs. 

Project Moonshot is also part of the move towards global testing standards. 

AI testing organisations – AI Verify Foundation (AIVF) and MLCommons – have signed a memorandum of intent to collaborate on building a common safety benchmark suite.