[Please check against delivery]
Let me begin by thanking the Ho Chi Minh City Department of Health and the Ho Chi Minh City Computer Association for hosting this conference on the important and exciting topic of artificial intelligence in health.
I am thrilled to be back in Ho Chi Minh City – just a couple of weeks after I was here last, for the amazing celebrations of the 50th anniversary of reunification. Congratulations to your city for hosting and organizing such a spectacular event.
It is a year of other big anniversaries: I understand this year also marks the 50th anniversary of the establishment of the Ho Chi Minh City Department of Health (congratulations!). It is also the 70th anniversary of Vietnamese Doctors' Day. WHO can also contribute a milestone to this impressive collection. We are proud to mark nearly 50 years of close cooperation with the Government and people of Viet Nam.
I mention all of this not just because each is worthy of celebration, but because anniversaries also provide a natural inflection point – for reflection on achievements, as well as consideration of the path forward with renewed perspective.
And this is why I am so honoured and happy to speak at this conference on AI in health care in Viet Nam, in Ho Chi Minh City – the heart of Viet Nam’s tech industry – and a city with ambitions to become a global leader in AI. Because AI – and digital technology more broadly – has the potential to revolutionize health care in the 21st century, in ways we cannot even begin to imagine yet.
At a national level, the Government – under the leadership of General Secretary To Lam – has put digital transformation at the heart of its reform and development agenda. I believe the health sector can be a driver of this transformation.
I know you will hear a lot over the course of your conference about the opportunities in AI for improving health and health care: here in Viet Nam, AI is already being used to analyze chest X-rays for multiple conditions, and optimize personalized treatment plans for different types of cancer, significantly improving accuracy and efficiency. Hospitals are increasingly adopting advanced robotic surgery systems to perform minimally invasive procedures with unprecedented precision, in fields ranging from general laparoscopy to complex neurosurgery. Together, these AI- and technology-driven advancements are elevating the quality of care, while optimizing the way hospitals are run.
But many other people will speak today about opportunities. So, I want to focus my remarks on challenges and risks: specifically, how we need to work together to address challenges, so we can harness the promise of AI for health, while mitigating the risks.
WHO’s Director-General, Dr Tedros (my boss), summarizes this as follows: “Like all new technology, artificial intelligence holds enormous potential for improving the health of millions of people around the world, but like all technology it can also be misused and cause harm.”
Let me briefly touch on two risks.
The first risk is that gaps in access to digital technology can exacerbate existing health and other inequities.
Across our Region, people in some places have rapidly adopted advanced AI – for example the Republic of Korea and Singapore. But take-up of technology is much lower where people are poorer, and have unreliable internet and low digital literacy.
Digital heath tools are only useful if we can access them, and we know how to use them.
For example, telemedicine has great potential to benefit a person with hypertension or diabetes living in a remote or hard-to-reach area, but only if that person actually has access to those possibilities – and the digital skills to be able to use them.
So, we need to guard against a future where the rollout of advances in AI and digital technology actually widens the equity gap between those with access to technology, and those without. We can mitigate this risk both by improving digital health literacy, as well as by ensuring the benefits of AI reach the people who need it the most.
When used wisely, AI can actually help to close gaps in access to health care.
For example, in remote areas of the Philippines, mobile health-care workers are using an AI-powered chest X-ray screening tool to identify TB cases, especially in vulnerable and high-risk populations. It also helps to identify immediate follow-up actions. As a result, more people are being diagnosed with TB, and fewer are stopping treatment, compared to traditional methods.
Thailand is rolling out an innovative web app that uses AI to analyse and evaluate patients' risk of stroke. This tool is designed to assist doctors in community and small hospitals, where there may be a lack of cardiologists, and enable quicker and more accurate decision-making – to protect people from disability caused by strokes which do not receive quick treatment, and prevent deaths.
A second area of risk is in the tricky area of data privacy and cybersecurity.
Without strong privacy and data security protections in place, people will not trust digital health – but building trust in digital health is fundamental to its adoption.
We will only be able to successfully implement digital health tools and tackle digital inequalities if patients, communities and health workers believe their data is safe and secure.
So, like every country trying to keep up with rapidly evolving digital technology, in Viet Nam we will need to balance privacy, security and national governance in the legal frameworks covering digital health and use of AI in the health system.
***
To ensure the potential of digital health for improving health can be realized, we need an enabling environment, grounded in strong governance – to implement the transformation ethically, safely, securely and equitably.
A successful digital health transformation will also require listening to the voices of patients and other service users, health-care workers, health leaders and policy makers across different agencies, and strong collaboration across sectors.
Balancing the opportunities and the risks is core to WHO’s work on AI in health care, which includes providing guidance, frameworks and partnerships to help ensure the ethical, equitable and safe use of AI in health systems.
In the past six months, the Government of Viet Nam has been developing a strong policy foundation to advance AI in health care. In December 2024, the Government issued a resolution that set the strategic direction for AI adoption, followed in January by a resolution detailing the implementation plan. Further solidifying this framework, the Ministry of Health launched its action plan in March, outlining concrete steps to integrate AI solutions nationwide. Complementing these efforts, the National Assembly passed a resolution in February, emphasizing the critical need for ethical AI governance and safeguarding patient safety. Together, these policies demonstrate Viet Nam's commitment to harnessing AI's potential while ensuring responsible and equitable health-care innovation.
As Ho Chi Minh City and other areas in Viet Nam look to scale up use of AI in health, I would like to briefly highlight six areas WHO recommends for regulation and specific focus, in order to manage risks like the ones I mentioned earlier:
We need to build trust through transparency: for example, by being open about how AI systems are created and used.
Apply Risk-Based Oversight, tailoring regulation to risk level—e.g., wellness apps vs. diagnostic AI.
Ensure Inclusive, Private Data: including through ensuring that data that is fed into AI models represents Viet Nam’s diverse population.
Keep Humans in the Loop: Health professionals must guide decisions, even with AI support.
Enable System Integration: Ensure data systems are connected and supported by strong internet infrastructure.
(as mentioned) Promote Collaboration: Engage regulators, patients, professionals, industry, and government to ensure ongoing compliance.
Let me end by saying that, in this complex, important and exciting agenda, WHO is honoured and proud to accompany Viet Nam, including through support for policy development, as well as promoting and facilitating dialogue on some of the more challenging issues I’ve just described – as I hope I am helping to do here today.
So thank you again to all of you for being here, and I’m looking forward to today’s discussions – and the many discussions that I am sure will follow after today – about how we guard against some of the perils of AI, while at the same time making the most of its promise for a safer, and healthier Ho Chi Minh City, and a healthier and safer Viet Nam.
Xin cảm ơn!