When a natural disaster strikes or a conflict erupts, aid organizations rush to help. Increasingly, they're armed not just with food and medicine, but with artificial intelligence. AI can predict disease outbreaks, optimize supply chains, and analyze data from crisis zones faster than ever before. But as this technology becomes embedded in humanitarian efforts, a critical question emerges: Who is governing its use, and who is accountable when things go wrong?
This is the central challenge explored in a new commentary, "Power, governance, and accountability in humanitarian artificial intelligence," published in The Lancet Digital Health. The authors, Patrick Vinck and Phuong Pham of Harvard Medical School, the Harvard T.H. Chan School of Public Health, and Mass General Brigham, are calling for a serious examination of the ethical frameworks surrounding AI in these high-stakes environments.
While the potential for AI to save lives is immense, the authors' focus on power and governance highlights the risks. Without clear rules and oversight, AI tools could reinforce existing biases, misinterpret data from vulnerable populations, or create new dependencies on the tech companies and experts who control them. The commentary implicitly asks us to consider:
- Who owns the data? Is it the affected community, the aid organization, or the tech provider?
- Whose values are coded into the algorithms? How can we ensure they align with humanitarian principles?
- What happens when an AI model makes a mistake? Who is responsible for the consequences in a chaotic crisis zone?
The authors' background as co-founders of KoboToolbox, a non-profit, open-source data collection platform used widely in humanitarian settings, gives their perspective unique weight. They have firsthand experience with the complexities of gathering and using data ethically in the field.
The Medicup Perspective
At Medicup, we believe that technology must empower, not overpower. The questions raised about humanitarian AI are directly relevant to all of digital health. Whether an AI is helping a doctor diagnose a condition via telehealth or assisting a relief worker in a crisis, the core principles of accountability, transparency, and patient-centricity must be the foundation.
Building robust governance isn't a barrier to innovation; it's what makes innovation sustainable and trustworthy. As we develop AI health tools, we are committed to creating systems that are not only effective but also equitable and accountable to the patients and providers who rely on them. This commentary is a vital reminder that with great technological power comes an even greater responsibility to the people we serve.
Source: [PubMed](https://pubmed.ncbi.nlm.nih.gov/42624703/)