Amazon Web Services seeks to usher in Singapore’s “AI Spring”

The new AWS AI Spring for Singapore programme will accelerate the adoption of AI in Singapore and contribute to digital economy goals, shared Tan Kiat How, Senior Minister of State at the Ministry of Communications and Information at the 2024 AWS Summit in Singapore. The new programme is spearheaded by AWS, who will be partnering with public and private sector players.

The AI Spring programme consists of six pillars which aim to drive AI adoption in the public sector and local enterprises, nurture AI startups and research and development, as well as provide skilling and professional certification programmes and contribute to community development with AI.

AWS is set to double its investment in Singapore, with newly announced plans to invest an additional S$12 billion (US$8.8 billion) in Singapore by 2028. The multinational hyperscaler has currently invested S$11.5 billion (US$8.5 billion) in Singapore, according to Priscilla Chong, Country Manager of AWS Singapore in her keynote speech.

AI Spring to complement the National AI Strategy 2.0

The newly announced AI Spring initiative will support Singapore’s digital economy goals, which currently consists of national level plans like the National AI Strategy 2.0, the Digital Connectivity Blueprint, and the upcoming Digital Enterprises Blueprint, shared Tan in his keynote.

As part of AI Spring for Singapore, AWS has announced two partnerships with the Infocomm Media Development Authority (IMDA), which oversees digital economy efforts in Singapore.

First, the new GenAI x Digital Leaders programme aims to support digitally mature enterprises develop and deploy their own generative AI solutions through workshops and access to AWS’ generative AI expertise and tools. They are targeting to support 100 enterprises via these workshops.

“We believe AI is a force for public good and a tremendous opportunity for Singapore and Singaporeans,” said Leong Der Yao, Assistant Chief Executive of Sectoral Transformation at IMDA, in a media briefing. Earlier this year, the agency launched a generative AI sandbox for SMEs.

Complementing IMDA’s efforts to upskill young talent, AWS has set itself a goal to train 5000 individuals from the Institute of Technical Education (ITE) and polytechnics in AI skills yearly over the next three years. They have signed an agreement with ITE to incorporate AI education into ITE’s curriculum.

AI will disrupt all industries, and even students enrolled in non-technical programmes stand to benefit from AI training, said Low Khah Gek, CEO of ITE, in the same media briefing. The institution is also training staff members in the use of AI.

Generative AI within the public service

Within government, generative AI can supplement the efforts of traditional AI to improve productivity gains for public officers, enhance core government capabilities, and innovate new services for citizens and businesses, shared Henry Chang, Deputy Chief Executive of Singapore’s Government Technology Office (GovTech), during the keynote presentations.

While traditional AI can support public officers in making more data-driven decisions, generative AI can help with generating content.

“GenAI’s ease of use and content creation capabilities have changed how we think about approaching complex tasks and data analysis,” he said.

For example, GovTech worked together with the Ministry of Manpower to develop a generative AI sensemaker tool built on Amazon Sagemaker and Bedrock, via GovTech's Analytics.Gov, to help policy officers better process large amounts of employment-related documents and feedback.

The tool uses clustering algorithms and a large language model (LLM) to extract insights and drive quicker operations and policy interventions, he said.

This has led to a 60 per cent improvement in their ability to extract insights, a 50 per cent reduction in time required for sensemaking, and saved 2000 hours across 3 months.

He noted that there are a growing number of use cases for generative AI in government, which includes intelligent chatbots, multilingual government communications, writing code, and fraud detection.

Some principles that inform GovTech’s approach to designing generative AI applications include optimising for cost, designing for flexibility, keeping humans in the loop, and maintaining data confidentiality and security, he said. The team is also investing efforts to address emerging threats, such as adversarial AI attacks and inference attacks.

GovTech’s generative AI efforts are supported by its history of government digitalisation. For instance, policy officers can request core government datasets through platforms like vault.gov and analytics.gov, and the government has successfully migrated more than 70 per cent of eligible workloads to the cloud as of 2023, he said.

This article was amended to highlight that GovTech and the Ministry of Manpower's new generative AI sensemaker was built via GovTech's Analytics.Gov, a whole-of-government data exploitation platform. You can read more about the project here and register to attend an upcoming webinar on how governments can deploy whole-of-government data analytics platforms here!