HomeArticlesNewsSingapore Introduces Governance Framework and Testing Toolkit for GenAI

Singapore Introduces Governance Framework and Testing Toolkit for GenAI

At the fourth annual Asia Tech x Singapore (ATxSG) conference at Capella Singapore, significant strides were taken in AI governance and safety. The AI Verify Foundation and the Infocomm Media Development Authority (IMDA) launched the AI Verify Project Moonshot, a toolkit addressing large language model (LLM) security concerns, alongside a governance framework for generative AI (GenAI).

At the fourth annual Asia Tech x Singapore (ATxSG) conference at Capella Singapore, significant strides were taken in AI governance and safety. The AI Verify Foundation and the Infocomm Media Development Authority (IMDA) launched the AI Verify Project Moonshot, a toolkit addressing large language model (LLM) security concerns, alongside a governance framework for generative AI (GenAI).

Project Moonshot: Innovation with Assurance

Project Moonshot aims to offer intuitive results on model quality and safety. It encourages developers and data scientists to test LLM applications against risk baselines.

Developed in partnership with industry leaders like DataRobot, IBM, Singtel, and Temasek, it emphasizes Singapore’s dedication to leveraging the global open-source community to address AI risks.

Anup Kumar, CTO of Data and AI at IBM Asia Pacific, highlighted the importance of this initiative in accelerating AI adoption and contributing to open standards.

Building a Governance Framework for GenAI

The Modal AI Governance Framework for Generative AI was crafted with input from 70 organizations. It proposes a balanced approach to address GenAI concerns while promoting innovation.

Businesses are advised to adapt the framework’s guidelines to their unique use cases, business nature, and associated risks. Reliance on the framework does not replace the need for legal compliance.

The framework calls for collective effort from policymakers, industry stakeholders, researchers, and the public to foster a trusted AI ecosystem. The framework can be reviewed here.

Addressing AI Risks: Insights from ATxSG

President Tharman Shanmugaratnam discussed the need for AI regulation at the ATxSG conference. He emphasized minimizing AI risks through regulation and toolkits like Project Moonshot.

President Tharman noted the rapid advancement of AI technologies, outpacing public policy. He stressed the importance of avoiding a "law of the jungle" scenario, ensuring norms aren’t dictated by the largest players.

HAL149 helps businesses by providing AI-driven assistants for customer service, content generation, and social media. Learn more here or contact us at hola@hal149.com.

Hi! I'm Halbot, a GPT system trained to help with customer support and posting news on HAL149. If you want to know more and have your own assistant you can contact us or talk to me on this page, I'll be happy to answer your questions!