Higher-level investments in platforms key to turn POCs into successful products

Infrastructure, access and capability must all move together, said Microsoft’s Managing Director for Singapore Chia Wee Luen, highlighting that national artificial intelligence (AI) readiness goes beyond implementing the tech itself.

He was speaking in the opening keynote at the Microsoft Public Sector Solutions Day on April 9 in Singapore.

In the panel “Making AI Work in Government: Culture, Trust, and Scale,” speakers from Singapore’s Central Provident Fund Board (CPFB) and Microsoft shared that the real value of innovation comes from enterprise-grade AI platforms and data accessibility, and not just local productivity tools.

CPFB’s Deputy Chief Executive (ICT & Digital Services) Ng Hock Keong was in the panel that was moderated by Microsoft ASEAN’s President Mayank Wadhwa.

The panel was followed by case study sharing by the National University of Singapore (NUS), Land Transport Authority (LTA) and JTC Corporation, on how they leveraged Microsoft’s products to scale their AI initiatives securely.

Here were the key takeaways from the opening sessions of the day.

1. Invest in shared platforms, not only one-off POCs

Assuming that most proof-of-concepts (POCs) technically work, the real question is whether they can scale and sustain at the enterprise level to deliver real impact, highlighted CPFB’s Ng.

The key takeaway was for government agencies to invest in secure enterprise-grade AI platforms and common data foundations.

Beyond the moral support from senior leadership to experiment with POCs, it is key for these platform investments to be made at the organisation level as “part of the cost of innovation,” said Ng.

“The platforms cannot be funded by any single project team. The return on investment (ROI) will not make sense.

“It has to be the internal investment that your C-suites agree to and put in – and even be prepared to write it off if the platform eventually doesn’t work,” he explained.

Wadhwa highlighted CPFB as a good example of moving from AI experimentation to real, scalable impact, with 99.5 per cent of CPFB’s transactions already conducted digitally.

“[Given] the role that CPFB plays at the national level where trust is non-negotiable, it has continued to build, modernise and innovate for decades,” he said.

2. Lead with trust and culture, not tech

As AI moves from talking (with natural language and GenAI models) to doing (with AI agents), the stakes around citizen’s trust and the governance layer will get higher.

Trust is non‑negotiable in public service, Wadhwa highlighted, especially when managing citizens’ money or sensitive data.

As AI continues to transform the workforce, the panel highlighted the need for top management to repeatedly and credibly signal that using AI is about supporting employees to better their work lives and improving citizen outcomes, rather than simply cost cutting.

"When it comes to the AI adoptions, there comes the million-dollar question of how we can use AI and yet not compromise the trust. This has been a very long discussion we had, even internally,” Ng explained.

AI success isn't just about the tech, but clear and consistent leadership messaging with a culture that champions change through small-scale pilots and active internal engagement.

3. Target AI where it frees public officers to do higher-value work

CPF officers use various AI tools in their digital platform to improve customer experience, manage their admin tasks and digital product development so they can focus on higher-value work, and are exploring customised AI agents to enhance customer service.

Ng believes that the next leap is when AI transacts and executes tasks on behalf of the citizens.

“For AI to add value, it needs to go out of government services alone, and be an enabler for the citizens in their day-to-day lives,” he said.

Most citizens want to work, live and play without the burden of interacting with public agencies, he noted. This is why Ng sees AI as a path to this frictionless reality.

To achieve this, robust engineering and a high degree of public trust in digital governance will be key.

As part of the key takeaways for the audience, Microsoft also announced several initiatives that will make it more secure and seamless for Singapore public officers to leverage data insights and AI capabilities on its platform.