Getting your Trinity Audio player ready...
|
In a collaborative effort, the AI Verify Foundation (AIVF) and the Infocomm Media Development Authority (IMDA) have introduced a draft Model AI Governance Framework specifically tailored for Generative AI, expanding upon the existing framework for Traditional AI last updated in 2020.
Generative AI holds transformative potential, surpassing the capabilities of Traditional AI, but it also introduces new risks. Recognising the need for consistent principles in this dynamic space, the proposed framework aims to establish a trusted environment for users to confidently and safely utilise Generative AI, fostering international conversations among policymakers, industry stakeholders, and the research community.
The framework addresses nine key dimensions to support a comprehensive and trusted AI ecosystem, emphasising core principles that decisions made by AI should be explainable, transparent, and fair. Beyond principles, it provides practical suggestions for model developers and policymakers, building on insights from a discussion paper on Generative AI and ongoing evaluations within the Generative AI Evaluation Sandbox.
AI governance is an evolving field, and international consensus is crucial. The proposed Model Governance Framework for Generative AI builds on the successful mapping and interoperability of national AI governance frameworks between Singapore and the US, setting the stage for further developments. The framework, available for review, invites feedback from the international community, with the finalisation expected in mid-2024.
For more details, the proposed draft framework can be accessed here, and feedback can be submitted to info@aiverify.sg to contribute to the framework’s completion.
As Singapore advances its digital economy, ensuring a trusted AI ecosystem becomes paramount, aligning with global AI ethics and governance standards. The Infocomm Media Development Authority (IMDA) had introduced AI Verify, a governance testing framework and toolkit designed to validate AI systems against 11 widely recognised ethics principles, fostering innovation while safeguarding consumer interests.
The 11 governance principles include transparency, explainability, safety, security, fairness, and more, forming the foundation for AI Verify’s testing processes, comprising technical tests and process checks. The integrated software toolkit operates within the user’s enterprise environment, enabling standardised tests on AI models and generating comprehensive reports, allowing companies to enhance transparency by sharing these reports with stakeholders.
While AI Verify presently focuses on common supervised-learning models, it does not cover Generative AI/LLMs, set ethical standards, or guarantee absolute safety. To bolster AI testing capabilities globally, IMDA establishes the AI Verify Foundation, a not-for-profit organisation fostering collaboration, contributing to AI testing frameworks, and driving broad adoption through education and outreach. The foundation seeks to harness the collective expertise of the global open-source community, advancing responsible AI use for companies and regulators worldwide.
The AI Verify Foundation (AIVF) harnesses the collective power and contributions of the global open-source community to develop AI testing tools enabling responsible AI. The Foundation promotes best practices and standards for AI and seeks to build trust through ethical AI.
The Infocomm Media Development Authority (IMDA) leads Singapore’s digital transformation by developing a vibrant digital economy and an inclusive digital society, fostering growth in Infocomm Technology and Media sectors in concert with progressive regulations, harnessing frontier technologies, and developing local talent and digital infrastructure ecosystems to establish Singapore as a digital metropolis.
Singapore demonstrates a keen interest in incorporating Artificial Intelligence (AI) and Generative AI across various fields and sectors, aligning with its commitment to stay at the forefront of technological advancements. Recognising the transformative potential of AI, the country aims to leverage it to enhance efficiency, innovation, and overall competitiveness in diverse industries.
OpenGov Asia reported that the Singapore Academy of Law (SAL) is actively strengthening the training and professional advancement of lawyers to elevate Singapore’s standing as a prominent legal services hub. In recognition of the increasing impact of infocomm technologies, especially generative artificial intelligence (AI), in the legal field, SAL is aligning its efforts with the recommendations of a Working Group entrusted with evaluating the legal education and training framework.