EY helps clients create long-term value for all stakeholders. Enabled by data and technology, our services and solutions provide trust through assurance and help clients transform, grow and operate.
At EY, our purpose is building a better working world. The insights and services we provide help to create long-term value for clients, people and society, and to build trust in the capital markets.
How government leaders can foster a thriving AI ecosystem
In this episode of the Government Insights podcast, host Kait Borsay discusses with her guests the role of governments in advancing AI ecosystems and contrasting their goals with business interests.
The artificial intelligence (AI) policy landscape is complex, with rapid technological changes and geopolitical factors influencing policymaking. How different organizations are leveraging AI also vary: Governments aim to foster AI innovation, develop skilled workforces, and ensure responsible AI use, while business use cases focus on AI for strategic cost-efficiency and profit growth.
Governments are considering various policy levers, from voluntary guidelines to adapting existing regulations, to stimulate AI innovation and manage risks. Our conversation emphasizes the importance of ethical AI governance, noting the importance of common principles like fairness, privacy, and accountability. A flexible, transparent, and internationally coordinated approach is proffered as a strategy to navigate the dual nature of AI as both a threat and an opportunity, ensuring a trustworthy AI future.
Balance innovation and responsibility by fostering AI innovation while ensuring ethical use, balancing technological advancement with societal values and public trust.
Invest in local education and infrastructure, such as computing resources and support for research and development, to capitalize on the potential that AI offers.
Collaborate internationally to establish common standards and ethical guidelines, ensuring global interoperability and the responsible AI evolution.
Teaser
Government Insights from EY-Parthenon.
Kait Borsay
Hello and welcome to Government Insights, a podcast series from EY-Parthenon for government leaders around the world.
I'm your host Kait Borsay, and we're looking at how governments can transform to strengthen services for their citizens.
In this episode our topic is how can governments foster a thriving artificial intelligence (AI) ecosystem. Joining us to share their insight and opinion on this subject are Anne McCormick, EY Global Digital Public Policy Leader. Hello Anne.
Anne McCormick
Hello Kait.
Borsay
Lovely to have you with us. And Gary Ong, Government and Public Sector Leader for EY-Parthenon at Ernst & Young Solutions LLP in Singapore. Hello Gary.
Gary Ong
Hi there.
Borsay
Gary, let's start with you. I wonder if you could tell us about some of the AI issues that governments and businesses are facing today.
Ong
Sure. Happy to do that. I think AI is sort of both an opportunity and a threat, right. So governments want to do broadly, I think four different things with AI.
Number one, they want to build it as sort of a thriving AI ecosystem. They want to develop their workforces right, with the skills and capabilities to capture the value added and employment opportunities presented by the technology. So that's number two.
And three, I think on the minds of governments is they want to seed the right infrastructure capacity to achieve their sort of national AI ambitions.
And last but certainly not least, I think all governments are sort of thinking about how to adopt responsible AI frameworks, how to foster like a trusted environment that protects users and facilitates innovation in AI.
Borsay
And how might it differ for business?
Ong
So for businesses, it's a little different. I think businesses look at it from the perspective of, you know, how can I leverage the technology to increase revenue or to reduce costs actually. And so with the way businesses have looked at it is from a pretty much from a use case perspective actually. So how do I apply the technology to business?
And if you see in examples of how the technology in particular, for more recent iterations like generative AI (GenAI) have been adopted, right, you'll see that in fact the use cases are actually really broad and very substantial. And I think that's what's interesting about the technology.
Borsay
And Anne, how does the playing field in terms of AI policy development look right now?
McCormick
I would say probably just starting with three words, complex, competitive and experimental. I mean really looking at the past 12 to 18 months, a huge amount has happened and a bit of a swing from the Chatbots, etc., euphoria to caution and perhaps a more focused and specific set of discussions around defining AI risk and opportunity.
I think today you know, I'd probably say two things to watch. First of all, policy makers just like business leaders are really challenged by the pace of change, shifting public expectations, let alone the difficulty of calibrating policy approaches to yet unknown use cases of really complex technologies, AI, but also connected technologies.
I think the second thing I'd say is obviously geopolitics are really shaping policy making and the tech race, but also geopolitical safety and security concerns. You know, on the one hand, you have policymakers assessing the perceived or misperceived tension I would say, between innovation and regulation. The desire to quickly develop competitive tech ecosystems. But also to address some of the societal concerns and risks that are coming into focus.
The other part of that is, paradoxically, also a recognition of the cross-border nature of technologies and of shared risks, is leading to a number of initiatives toward international coordination, including between geopolitical rivals. So really kind of a complex dynamic environment.
But a lot of learnings coming through, a lot to copy, to adapt, to innovate from a policy perspective as well as from a business perspective.
Borsay
Gary let's focus more with you then on cross-border AI and governments working together and I wonder what the development priority is that governments around the world are, what do they have in common?
Ong
I think different governments are kind of investigating how to best apply AI in the public sector in the delivery of public services. In Singapore, for example, one of those major examples has been to develop this app called One Service. One of the difficulties, for example, that citizens have, you know, engaging government, is that government is complex. You don't know who to approach to resolve a typical sort of government citizen products or services, right?
And what the AI enabled app allows you to do is query it in simple natural language. And the app is smart enough to direct you to the right agency, deliver the response if needed, query the right databases, get you the answer that you need, or direct it to the right agency for follow up, and that's brilliant. It simplifies what would typically be a very cumbersome process.
And so thinking more broadly into specific priorities within government services, for instance, transportation and mobility infrastructure, the technology to be deployed for even better traffic optimization. Predictive maintenance. We've always talked about smart cities, Kait, but now this is the potential to be really smart in a way that wasn't before actually, if you think about it.
Human Services, can we think about how to optimize welfare benefits distribution? In the area of justice, can we use it to improve significantly video analytics-based crime management, for instance. In education, think about customized learning to a degree that wasn't possible before. Think about the pain of grading and how AI could help to take that away, now that's brilliant.
In the area of national security and defense, how, for example, governments use the technology to kind of identify cyber-attacks and actively sort of defend against it. In terms of budget optimization for public finance management.
In terms of public health and how to kind of optimize healthcare operations, deliver personalized services. Now these are some of the broader priorities that the technology can potentially be used to really improve how governments actually deliver public services.
Borsay
That's very interesting. Anne, let's talk about some of the policy levers and strategies really that governments are using now to shape and support the AI space.
McCormick
Through to adapting existing regulations, for example, sectorial regulations in the UK and through to the other end of the spectrum, you know, with new targeted or really comprehensive laws with teeth, with severe penalties. And in some cases with extraterritorial and even geostrategic ambition, you know, like the EU's AI Act, trying to apply the Brussels effect a bit like the GDPR did.
Policymakers are also assessing the related governance disclosures and reporting aspects, the proof, whether it's voluntary reporting or mandated reporting. And the role of third-party oversight and independent verification. So at a macro level you can look at these.
I mean, the key thing for us is just to say that all of this is generally to be considered as part of promoting an ecosystem that meets policy objectives. So it will be a mix of levers and mix of policies. Looking at innovation levers there are a number of actions that we're seeing around the world. For example, governments widening access to key innovation building blocks.
For example, offering access to state owned supercomputing capacity, like in France, Germany, or government buying expensive AI chips and making them accessible to startups as public assets. I know the Government of India is looking into this at the moment. Investing in cloud computing, Singapore has been looking into this.
But also governments developing AI innovation sandboxes to support what we call kind of, bleeding edge innovation, while offering some safeguards and support on oversight.
Another lever is kind of encouraging self-regulation and co- regulation to build trust and adoption in AI, but also set out some expectations. So on the one hand, you want consumers and businesses to use AI fast. On the other hand, you want to make sure it's trustworthy, it's high quality. And if self-regulatory codes don't work, perhaps governments are also sending signals to what are their expectations and how those expectations eventually could also be translated in hard law if need be.
The third lever on innovation I would focus on is talent. You know there's a real need to develop and promote talent. And that kind of links with academic institutions and universities, but also, you know, visa and immigration policies to attract and retain the right talent. For example, the US recent executive order on AI really touches on that.
Turning to managing risks, kind of some of the levers are around making sure that there are common definitions of the core terms. You know, what is a risky AI system that may or may not require some oversight.
So in conclusion, it's really about an ecosystem-wide-approach using these different levers in a way that reflects particular government’s concerns and how they see the opportunities to be competitive internationally.
Borsay
Thank you. Let's look at examples, then that both of you have of how governments are experimenting with AI policy making to accelerate the development of thriving AI ecosystems. Gary?
Ong
When we use the term AI policymaking, actually it covers a broad swath right, of different initiatives the government is undertaking to kind of enable this ecosystem actually to take place.
And one of those examples is actually Singapore, which I'm very familiar with. I think the Singapore government recently published a National AI Strategy 2.0 an update to a version that they published quite a few years ago actually, they were one of the first to do it globally, and it really sets out a macro view really on what are the different areas that governments can intervene with policy to kind of drive the development of an ecosystem.
It can be looked at in terms of different industry levels, in terms of, you know whether government funding can be used to support, for instance, the development of centers of excellence for specific sectors in the economy around AI.
Government itself actually can get in and develop AI propositions to drive public sector productivity, as I mentioned earlier. And government with its huge funding mechanism, can actually be key in terms of driving research in the space and ensuring the right types between academia and industry, so that you know, the right kind of levers for continued development in the space are there.
Borsay
Anne, anything else from you on this? More examples, any other relevant examples really of how governments are experimenting with AI policy making to accelerate the development of thriving AI ecosystems?
McCormick
Yes. Well, just two quick ones which I love. I mean Japan, really focused on large tech industry including its strengths in automation and robotics and looking to modernize in parallel the chip manufacturing industry to meet their AI needs.
What's interesting is Japan has really focused on voluntary guidelines with buy in from the industrial base. So that's a really interesting and intentional approach.
There's also been internationally strong engagement and frankly leadership of the G7 on the Hiroshima principles and codes of conduct.
The government is also really interested in evaluating how and where to engage with the development of technical standards around AI development. As I alluded to earlier, these are absolutely fundamental to be specific about what's expected.
The UK, another interesting example, focus on applying existing regulations in often cases you know, sectorial regulations avoiding the need to create new regulatory requirements.
They're very engaged and in some cases, co-leading international efforts on AI and leading encouraging the development of national AI safety institutes which are there for developers and other AI players to consult on the safety of their systems and also for governments to learn and get a better sense of what's happening.
The UK is also supporting the AI startup environment through links with strong academic and AI research centers. So really kind of two interesting examples there.
Borsay
Finally, when it comes to trust and fostering a more ethical future, should we view AI as a threat or an opportunity? Anne?
McCormick
Ah, it depends. So innovate, think of the ecosystem and the links with technology beyond AI. So quantum look ahead, be flexible. But, but, but, ensure transparency and accountability.
Consider how you can encourage AI trustworthy kite marks. So how does aligning and being compliant with a voluntary code or a law, how can that be documented? You know, maybe with some third-party certification or a conformity assurance to say that your AI system or your use of this AI system is trustworthy and fantastic. I think that's a really interesting space. Be forward facing, this is going to need flexibility and really think international. This is a global playing field.
Borsay
And Gary?
Ong
I said at the outset, actually, that AI is both a threat and an opportunity, and I think a lot of governments look at it that way. It's an interesting question you've asked because we actually supported one of the countries here in ASEAN actually two years ago to develop principles on AI governance and ethics to develop and deploy AI responsibly.
So we've got quite a good view on this issue actually. We've actually helped define and governments are obviously all applying these sorts of ethical frameworks for their national economies in their own ways and specific ways that are more relevant to their stage of development. But we think these key principles for responsible, ethical adoption of AI can be fairly generalized into kind of maybe a few key points, including fairness and equity, data privacy, transparency, reliability and safety, and then accountability and oversight.
So I think these are some of the broad principles that were developed in our thinking that can be applied across the different governments in the region.
Borsay
Well, look, that is all for this episode. Such an enlightening conversation as well. Gary, thank you.
Ong
Thanks very much.
Borsay
And Anne, thank you to you.
McCormick
Thank you very much.
Borsay
Do join us again soon when we'll continue to look at how governments can transform to strengthen services for their citizens.
And please subscribe to this series so you won't miss an episode. From me, Kait Borsay, thanks for listening and bye for now.
Teaser
Government Insights from EY-Parthenon, back soon.
End of podcast.
Presenters
Kait Borsay
Journalist, author, TV presenter, Radio moderator at Times Radio
