peshkov -

Singapore government teams up with Google Cloud on AI

The Singapore government and Google Cloud will make AI capabilities, including developer tools and AI models, available to public sector agencies through a locally hosted AI platform

The Singapore government has teamed up with Google Cloud to make artificial intelligence (AI) capabilities available to public sector agencies in the city-state through a new AI cloud platform.

Dubbed AI Government Cloud Cluster (AGCC), the platform leverages Google Cloud’s Vertex AI technology stack that includes Nvidia’s A100 GPUs, AI developer tools, explainable AI and data governance toolkits, as well as a repository of pre-trained generative AI models.

Speaking at the Google Cloud Summit in Singapore today, Chan Cheow Hoe, the government’s chief digital technology officer, said discussions with Google to develop AGCC started in February this year as part of efforts to build a foundation for AI for Singapore.

That includes making first-party, third-party, and open-source AI models available to public sector agencies as well as the private sector through a “model garden”, Chan said.

Ting Liu, vice-president and principal software engineer for AI at Google Cloud, said with a model garden, developers will be able to discover models and launch their own AI projects.

“We not only offer state-of-the-art Google foundation models, we also bring in third-party partner models and also the very popular open source models,” Liu said, adding that developers can also use Google’s Generative AI Studio to test the models in a low-code and no-code environment.

The Government Technology Agency (GovTech) is the first public sector organisation in Singapore to leverage the AGCC, specifically the Pathways Language Model (Palm) on Vertex AI to power an assistive chatbot that civil servants can use to boost their productivity.

The Monetary Authority of Singapore, the city-state’s central bank and financial industry regulator, is also expected to use AGCC to customise other foundation models from Google in use cases such as anti-money laundering, regulatory compliance, as well as analysis of environmental, social, and governance data.

The AGCC, slated to go live in June 2023, is hosted in a dedicated and localised cloud computing environment in Singapore. The Singapore government and Google Cloud have also put in place privacy and security safeguards to maintain the confidentiality of government information.

The move by Google Cloud to make its AI services available to Singapore’s public sector agencies comes on the heels of efforts by major hyperscalers to win over influential customers and participate in cloud initiatives such as GovTech’s Government Commercial Cloud (GCC).

In 2021, GovTech said it was working on GCC 2.0 to make it easier for government agencies to manage and secure their use of public cloud services from the likes of Amazon Web Services, Microsoft Azure, as well as Google Cloud.

To address the AI talent crunch, Google Cloud will also partner with GovTech to design and run whole-of-government training programmes to help agencies build in-house data science and AI expertise, develop AI innovation strategies, and implement data governance best practices.

The training programmes are an expansion of Google Cloud’s existing public sector skilling commitments and will be offered in various tailored forms to 150,000 public officers across 16 ministries and more than 50 statutory boards.

Karan Bajwa, vice-president for Asia-Pacific at Google Cloud, said the cloud supplier will focus on building AI skills at all levels across Singapore, beyond the public sector.

“As we take you through the skills issue, we thought of various personas – not just IT managers, but fundamentally all personas. How do we take basics of AI to citizens? How do we look at government personas, how do we look at developers, startups and going all the way up to the higher-order skillsets.

“We’ve created multiple personas that we'll be working with the government on and bringing skills into the ecosystem with one thing in mind – scale and sustainability,” he said.

Read more about cloud in APAC

  • Melbourne-based Cortical Labs’ lab grown neurons could speed up AI training in a more energy efficient way and its work has caught the eye of hyperscalers and Amazon’s CTO.
  • India’s Cropin, one of the first movers in agriculture technology, has built an industry cloud platform with AI capabilities that is now used by the likes of PepsiCo to maximise crop yields.
  • SKT plans to broaden the use of AI across its business, from delivering AI-powered services to improving customer experience using generative AI models.
  • An AI engine developed by Singapore startup EntoVerse is helping cricket farmers improve yield by optimising environmental and other conditions.

Read more on Cloud computing services

Data Center
Data Management