“AI for impact” is thriving in Asia – so how can public systems speed up adoption?

AI is having a profound impact on the world, particularly in Asia, where social entrepreneurs are leveraging the technology to create initiatives that improve health services, strengthen food systems, expand access to education, support climate resilience and optimize the delivery of public services.

The very nature of social innovation means that entrepreneurs in this diverse and rapidly growing ecosystem are able to convert emerging tools into practical solutions for communities that are often hardest to reach, testing what works in real‑world conditions, identifying risks that need to be managed, and highlighting opportunities for positive impact. The potential is enormous.

In the Philippines, for example, a partnership between Philippine Business for Social Progress and Siemens Healthineers uses an AI tool to read chest X‑rays for signs of tuberculosis, enabling doctors to screen more people every day than they could alone.

Elsewhere in the country, the Institute for Climate and Sustainable Cities (ICSC) uses AI to map the hidden rooftop solar economy via machine learning and satellite imagery. By providing accurate data to city officials and the Department of Energy, the initiative helps formalize thousands of unregistered systems, expands incentives, and accelerates clean energy adoption across an urban centre of half a million residents.

And in Bangladesh, the water enterprise Drinkwell works with every public water utility in the country to run a network of 300+ water ATMs, reaching communities the piped network doesn’t serve. Drinkwell.AI is a utility co-pilot trained on 13 years of engineering experience: system designs, maintenance procedures, water-quality records and field troubleshooting. An operator who hits a fault can ask in plain language and get an answer in seconds, drawn from everything the team has already solved, instead of waiting on the one colleague who has seen the problem before. The result is less downtime on a network that has dispensed more than two billion litres of water.

These solutions are powerful because social innovators offer a bridge between technology and communities. Their on-the-ground knowledge and understanding of local communities means they can deliver solutions that are meaningful and effective.

However, despite thousands of promising AI-for-impact initiatives emerging across the region – for there is no shortage of ideas or ambition – many still struggle to move beyond experimentation and into tangible delivery. Despite clear demand, why aren’t more initiatives like these already being used in the public sphere?

A report from the Schwab Foundation, AI for Impact: The role of AI in social innovation in Asia, draws on a dataset of more than 2,800 AI‑for‑impact initiatives across 10 economies and finds that the answer is certainly not down to a lack of technical excellence or model sophistication. Instead, the lack of progress is down to ecosystem readiness, including issues around governance, financing, institutional capacity and trust.

Promising tools often fail to become embedded in public systems or frontline delivery simply because ecosystems aren’t ready for them. Innovators are hindered by poor data interoperability, institutional fragmentation, a lack of funding and funding biases that favour incumbent providers, which limits opportunities for new solutions.

In contrast, successful solutions tend to flourish in places where governments align strategy, infrastructure and partnerships, enabling deployment at scale. In Viet Nam, for example, adoption is accelerated through a comprehensive national AI strategy, three national AI innovation centres, open data development and the system-wide integration of AI into public services. China, Malaysia and Thailand similarly benefit from national policy, dedicated coordinating institutions and sustained public investment. These factors are critical for AI deployment at national scale.

AI‑for‑impact activity in the region is expected to continue expanding over the next three to five years. But without supportive ecosystems, many life-changing – and indeed life-saving – ideas will struggle to fulfil their potential. In order to move beyond experimentation and into implementation at scale, policy, infrastructure, talent, finance and partnerships need to be aligned around clear public interest outcomes and community priorities. In other words, we need to build the “institutional plumbing”.

For governments, this means publicly demonstrating support for AI solutions through appropriate investment and the creation of shared standards for data quality, privacy and interoperability. Digital public infrastructure and responsible data governance frameworks provide the foundations for proven AI solutions to plug into core systems such as health, agriculture and social protection, and allow social innovators to build once and deploy across multiple programmes.

Crucially, governments also need to create clear pathways for adoption. This includes procurement processes that prioritize measurable outcomes such as reduced waiting times or improved access to services; regulatory sandboxes that allow solutions to be tested directly within public systems; and stronger coordination across ministries so that solutions developed in one area can be integrated and scaled nationally. Viet Nam’s national AI strategy, for example, combines dedicated funding with sandbox regimes and mandatory AI integration across essential public services.

Technology companies have an opportunity to design products, platforms and business models with social innovators in mind. This includes developing accessible tools, modular architectures and pricing models that lower the cost and complexity of deploying AI in low‑resource settings, allowing smaller organizations to build on enterprise‑grade infrastructure. Service providers – such as consulting and professional services firms – can help bridge the gap between technology development and real-world application by supporting capacity building, skills development and partnerships between social innovators and the public sector.

Meanwhile, philanthropic organizations, development agencies and impact investors can drive further adoption by shifting capital from isolated pilots to ecosystem enablers. This includes funding for shared infrastructure, data collaborations, capacity‑building and multi‑year scaling. In Pakistan, for example, an AI‑enabled early-warning system is being scaled through coordinated investment in weather stations, data platforms and national infrastructure. It highlights that funding underlying systems is just as critical as investing in standalone tools to reach millions.

The acceptance of AI, and ultimately its spread across societies, will depend on whether it delivers clear, tangible benefits in people’s daily lives. Across Asia, that connection is being made not by technology alone, but by social innovators who translate abstract capability into concrete benefit – whether by improving access to healthcare, strengthening livelihoods or expanding essential services.

AI scales not through technical sophistication alone, but when it is embedded into the systems people rely on and designed around their needs. Where solutions are rooted in community context, integrated into public services and supported by the right infrastructure and partnerships, they can begin to move beyond pilots.

Disclaimer: The views reflected in this article are the views of the author and do not necessarily reflect the views of the global EY organization or its member firms.