Blitz Bureau
NEW DELHI: In January this year, over 30 inter national experts in Artificial Intel ligence, mental health, ethics, and public policy gathered for an online workshop organised by the Delft Digital Ethics Centre (DDEC) at the Delft Uni versity of Technology (TU Delft) – the first WHO Collaborating Centre on AI for health governance, including ethics.
Held as an official pre-summit event of the India AI Impact Summit 2026, with support from the World Health Or ganisation, the workshop convened re searchers, policy-makers, clinicians, and advocates. Dr Alain Labrique, Director of WHO’s Department of Data, Digital Health, Analytics and AI, noted: “As AI increasingly interacts with people in mo ments of emotional vulnerability, we as WHO and its stakeholders must ensure these systems are designed and gov erned with safety, accountability and hu man well-being at their core.”
Generative tools Central among these challenges is the growing use of generative AI tools – neither designed nor tested for mental health – for emotional support, particu larly by young people, and the potentially serious risks this may pose. “We are at a critical juncture”, Sameer Pujari, WHO’s AI Lead, remarked. “The pace of AI adoption in people’s daily lives has far outstripped investment in understand ing its impact on mental health. Closing that gap requires coordinated action and dedicated resources from both the public and private sectors.”
Underscoring the importance of cross disciplinary collaboration, Dr Kenneth Carswell of WHO’s Department of Non communicable Diseases and Mental For mental health and wellbeing Health added: “Minimising risks from gen erative AI for mental health while maxi mising benefits requires bringing together the voices of those most affected, clinical and research expertise, governance and regulatory frameworks, and data to inform understanding. WHO is committed to en suring that users’ wellbeing stays at the centre as these tools evolve.”
tural, linguistic, and contextual factors. The participants emphasised the im portance of consumer empowerment, while TU Delft’s Dr Caroline Figueroa highlighted the urgent need for consen sus on crisis referral frameworks and accountability systems.
A strategic pillar More broadly, the workshop illustrat ed how the WHO Collaborating Centre mechanism has become a critical pillar in implementing the WHO’s vision for re sponsible AI in health.
Through this mechanism, WHO mobil ises world-class academic expertise and convenes diverse international stakehold ers to generate evidence-based recom mendations in support of its standard setting role. As Dr Stefan Buijsman, Managing Director of the DDEC, noted: “As a WHO Collaborating Centre, we can increase impact by collaborating with ex perts around the world, domain experts, and governments.”
Looking ahead A major challenge is the growing use of generative AI tools for emotional support, particularly by young people, and the potentially serious risks this may pose Key recommendations The workshop distilled these dis cussions into three principal rec ommendations: First, generative AI use should be recognised as a public mental health concern, with commensurate re sponses across Government, health systems, and industry that address all generative AI solutions, not only those intended for mental health.
Secondly, mental health should be integrated into impact assessments and monitoring of AI solutions to better understand their effects on determinants of health, short-term clini cal measures, and long term outcomes, such as emotional dependence.
One workshop participant stressed: “We need independent in vestments to test these effects.” Thirdly, AI tools used for mental health support should be co-designed with mental health experts and people with lived experience, including youth. Tools must be grounded in the best available evidence and tailored to cul The WHO is establishing a Consortium of Collaborat ing Centres on AI for Health, a network of leading institu tions across all six WHO regions, to support Member States in the responsible adoption of AI.
A pre-convening of can didate consortium members took place on March 17-19 at TU Delft, where institutions aligned on shared pri orities and agreed on initial collaboration mechanisms to build the collaborative infrastructure needed to ensure that AI governance in health is grounded in evi dence, ethics, and the needs of diverse populations worldwide.
