Home/About Us

About Us

The Singapore AI Safety Institute (AISI) will drive Singapore’s AI evaluation and testing work, to support addressing gaps in global AI safety sciences.

It is crucial for the state of sciences for safety and governance to keep in tandem with AI’s rapidly evolving capabilities and risks. As the designated Singapore AISI, the Digital Trust Centre Singapore (DTC), in partnership with the Infocomm Media Development Authority of Singapore (IMDA), will drive the efforts for advancing AI safety in Singapore.

Our Work

 

As AI systems become more capable, autonomous, and deeply embedded in society, it is essential to ensure that the development and deployment of AI is safe, reliable and trustworthy. Our work is dedicated to advancing AI governance and safety through the following core areas, to help organisations, policymakers and researchers navigate associated risks while optimising the benefits of AI:

 

Technical Testing

We design and conduct technical testing to develop evaluation tools and common best practices, as well as to understand AI risks better and improve testing methodologies. Our testing work includes bilateral and joint testing efforts with other members of the international Network for Advanced AI Measurement, Evaluation and Science (NAAIMES), and focuses on two key areas:

 

  1. Multilingual, multicultural testing: To ensure AI works reliably, fairly and safely across languages, norms and lived experiences. It helps uncover failures that only appear in non-English settings, reduce risks of biased or culturally insensitive outputs, and make sure safety measures uphold for even minority groups.
    As part of efforts to uplift the efficacy of model evaluations across languages, we conducted testing spanning across topics such as multilingual agentic testing using our open source evaluation tool Moonshot (as part of NAAIMES’ joint testing efforts), multilingual translation of local safety benchmarks (conducted with Japan AI Safety Institute), and an evaluation of LLM judge consistency across languages (conducted with Canada AI Safety Institute).
  2. Agentic AI testing: As AI systems become more autonomous and goal-directed, we study how agentic behaviours emerge and scale to provide a better understanding of associated risks. Our evaluations include data leakage risks in routine agentic tasks, which we conducted with Korea AI Safety Institute.

Societal Impact Studies

A key focus area is studying how AI systems affect individuals, institutions, and broader societal structures. Our work examines issues such as emotional dependency and cognitive offloading, ensuring that AI deployment aligns with public interest and long-term societal well-being.

 

By integrating technical rigor with policy insight, we aim to shape a future where AI systems are not only powerful, but safe, accountable, and aligned with human values.

International Partnerships

We recognise that AI governance and safety should be a globally coordinated effort. Meaningful progress in global AI safety sciences depends on collaboration across countries, institutions and communities to share expertise and build globally relevant governance approach and methodologies.

 

Regionally, we work closely with partners in ASEAN and Asia-Pacific to build capacity in testing and support the understanding of how associated AI risks can manifest across different scenarios and linguistic contexts. This is done through the workshops and red teaming challenges that we have conducted for knowledge sharing and capacity building in technical testing capabilities.

 

Internationally, we collaborate on joint testing exercises with members of NAAIMES, and continue to align with global priorities in AI governance and safety through contributions in events such as the International Scientific Exchange on AI Safety.

 

View our work here.

 

 

 

 

 

Who We Are

The Infocomm Media Development Authority of Singapore (IMDA)

IMDA is the overall government agency for the Singapore AISI, responsible for developing baseline policy, as well as fronting international engagements and partnerships related to AI safety with other AISIs and government agencies. IMDA also partners the Digital Trust Centre (DTC) to drive the technical work for the Singapore AISI.

 

As a statutory board in the Singapore government, IMDA leads Singapore’s digital transformation by developing a vibrant digital economy and an inclusive digital society. As Architects of Singapore’s Digital Future, we foster 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.

 

Find out more about IMDA here.

The Digital Trust Centre (DTC)

Hosted in the Nanyang Technological University (NTU), the Digital Trust Centre is a national centre that leads Singapore’s research and development efforts in trust technologies, including AI safety. The objectives of the DTC, which aligns with those of the Singapore AISI, are to pull together Singapore’s research ecosystem to enable industry digitalisation and to build a strong core of talent to strengthen Singapore’s status as a trusted hub in the digital economy. DTC’s work is funded by a $50 million grant from the National Research Foundation, Singapore (NRF) and the Infocomm Media Development Authority of Singapore (IMDA), starting from 1 October 2022.

 

The DTC will focus on:

  • Trust Tech Research: Enabling Institutes of Higher Learning and Research Institutes to pursue research excellence and drive collaborations.
  • Trust Tech Innovations: Encouraging academia and enterprises to co-develop market-ready solutions.
  • Sandbox Environment: Allowing businesses to experiment with Trust Technologies for data sharing challenges.
  • Local Capabilities: Nurturing 100 R&D talents in digital trust.

Read more about DTC here.