양리밍 장관의 국가 슈퍼컴퓨팅센터(NSCC) ASPIRE 2B 슈퍼컴퓨터 출범식 개회 연설

Professor Tan Chor Chuan

Mr Quek Gim Pew

Dr Terence Hung

Colleagues and friends

Good afternoon and thank you for inviting me.

Ten years ago, ASPIRE 1 came online as our first national supercomputer. Many of you helped to make it happen.

Your efforts have made it possible for the research community to access high performance computing resources to carry out important work. This access remains essential, particularly in the context of our aspirations for AI.

As you have heard me say on several occasions, we have made good progress since the National AI Strategy 2.0 was launched about 30 months ago.

Over 70 companies have set up Centres of Excellence here and are working with the local AI community to deepen their capabilities in Singapore.

Frontier labs like Google DeepMind and Microsoft Research Asia, and emerging start-ups like AMI Labs and Cognition, have established regional teams here. They are putting AI to work in industry and pushing at the hard research questions.

Our AI-for-Science effort is tackling pressing challenges like vaccine development.

These developments are encouraging us to activate every part of our AI ecosystem to harness AI for the Public Good, for Singapore and the World.

Just last month, we refreshed our priorities across the 10 enablers identified in the National AI Strategy. We will be supporting the next bound of Singapore’s AI efforts in three main ways:

First, going deeper in transforming key sectors of our economy;

Second, making AI adoption the norm rather than the exception; and

Third, strengthening Singapore as a vibrant AI hub.

We established the National AI Council, chaired by Prime Minister Lawrence Wong, to provide strategic direction and advance our AI agenda.

The Council will steer our national AI Missions in four key economic pillars – connectivity, advanced manufacturing, healthcare, and finance- which together contribute over 40% of our GDP.

It will also provide targeted support to organisations that can be Singapore’s “Champions of AI”.

Research remains a cornerstone of our ambitions.

We are committed to research about AI, to understand AI better, build new tools, and allow Singapore to remain at the forefront of AI.

We are equally vested in research using AI. We see AI as a multiplier for our research community, to redesign your workflows and speed up breakthrough discoveries and applications.

To support these aims, we have committed over S$1 billion to fund public research into fundamental and applied AI, and to develop AI talent in R&D.

At the same time, we will enhance our capabilities in advanced computing.

In Singapore, NSCC supports a core community of over 700 active users, and a wider group of around 9,000, across our Institutes of Higher Learning.

In the relatively short time since 2024, NSCC has enabled over 1,500 projects, delivering impact across research and industry applications.

For example, with NSCC’s support, homegrown maritime engineering specialist Mencast built a platform for designing marine propellers.

The platform combines AI-driven design, optimisation and high-fidelity simulations to significantly shorten development cycles.

In a matter of days, engineers produced and explored more than 10,000 design iterations; previously, they would have taken weeks just to produce 20.

Today, with the ASPIRE 2B launch, many more companies like Mencast can benefit.

ASPIRE 2B will bring about a 100-fold increase in compute capacity as compared to ASPIRE 1.

It will have more than 1,500 Nvidia H200 GPUs and can deliver up to 115 petaFLOPs.

This is of course nowhere near the cluster sizes available to frontier model developers, but significantly more than the clusters usually available for academic and industrial R&D.

In any case, rather than focusing on the numbers, we should be looking at the scope and substance of our experimentations going forward.

With ASPIRE 2B, models that were previously too large, can now be trained in Singapore to meet our specific needs.

Simulations that ran as approximations can now run at resolution.

Workloads that had to be sent overseas, can now use our national research infrastructure.

Let me give three specific examples of the possibilities.

The first is climate - an area of vulnerability for island states like Singapore.

Drawing on compute from NSCC, our Third National Climate Change Study produced high-resolution models tailored to our urban environment. These models were very helpful for designing our coastal protection strategies and guiding our longer-term investment decisions under the Singapore Green Plan 2030.

With ASPIRE 2B, we can go further. Researchers can develop advanced climate modelling approaches that combine AI and physics-based simulations, for higher resolution forecasting, sharper and more granular climate insights.

This can help us anticipate intense rainfall and rising seas earlier, and plan our urban development and coastal defences around them.

The second is in ensuring that our region’s languages, cultures and contexts are understood and represented.

You heard earlier from Dr Terence Hung who spoke about SEA-LION and subsequently, MERaLion — built by A*STAR, is the world’s first multimodal model that understands Southeast Asian languages, cultures and nuances.

Something that people may not realise is that Southeast Asia is rich in languages and cultures. There are in fact more than 1,000 languages spoken in our region. So, to be able to have these reflected in models that our companies and entities operating in our region use, is something valuable.

The third is still ahead of us. As Dr Hung mentioned, NSCC will integrate a quantum computer to ASPIRE 2B this year.

Pairing quantum and classical systems is the start of a new journey for us.

We don't fully know what it can unlock. But our researchers can start to explore compute-intensive areas like molecular simulation and advanced materials.

This will position us well to experiment and harness the benefits of quantum computers.

Beyond compute infrastructure, we must build human capabilities.

Even without quantum computers, next-generation AI technologies, such as agentic and physical AI, will involve significantly more computational intensity than systems today.

NSCC should expect more requests for support. But as Professor Tan Chor Chuan put it, it is not just about having more capacity. It is also about having the capabilities to meet those requests.

Therefore, I am very glad to hear Dr Hung speak about NSCC’s intention to focus on user enablement, and also to empower your users to make full use of the infrastructure available to them.

The NSCC team will:

Increasingly be tasked to manage complex compute demands and infrastructure;

Be asked to deliver streamlined services where researchers have timely access to integrated AI, simulation and computing capabilities; and

We look forward to seeing it equip our students and research community with the skills to use advanced compute for research and innovation.

Our measure of success cannot be limited to the scale of hardware.

What matters is how well we use the infrastructure — efficiently, effectively, and in service of the research community.

People and skills will still be essential to turn hardware accessibility into robust climate models, medicines, or better tools for industry.

I therefore strongly support NSCC’s focus in parallel to scaling its impact by:

Rethinking how it engages the research community;

Finding new ways to collaborate and enable innovation; and

Leaning forward to anticipate and support emerging needs.

Over the past decade, NSCC has been an indispensable partner to our research ecosystem.

I thank the NSCC team for its contributions and look forward to its greater success.

Thank you.