Announcing the Lancet Global Health Commission on artificial intelligence (AI) and HIV: leveraging AI for equitable and sustainable impact
Read the article
Despite substantial progress towards achieving the UNAIDS 95–95–95 targets,1 the global HIV response continues to be undermined by a multitude of complex, interrelated challenges. Persistent stigma and discrimination deter many from seeking testing and treatment.2 Socioeconomic inequities also limit access to health care in many settings, and in sub-Saharan Africa, women account for nearly 60% of new infections due to gender-based disparities.1 Furthermore, inadequate health-care infrastructure in high-burden, low-income regions hampers effective prevention and treatment efforts. Furthermore, inadequate health-care infrastructure in high-burden, low-income regions hampers effective prevention and treatment efforts. These challenges are exacerbated by scarce domestic resources, stagnating donor contributions, and shifting geopolitical dynamics, including the recent strategic re-calibration of US global health engagement, which threatens the sustainability of critical health programmes.
In the context of these challenges, integrating artificial intelligence (AI) into HIV programming could enable the transformations necessary to realise the UNAIDS targets by leapfrogging many of the technical and programmatic barriers that have slowed progress within the past 10 years.3 Already, AI tools are enabling more responsive public health responses through the use of predictive analytics;4,5 in Kenya, AI tools have been deployed in 1800 health-care facilities to more precisely identify populations at risk of HIV acquisition, optimising case finding and enabling personalised prevention strategies.6 Additionally, AI-enabled geospatial mapping is pinpointing high-risk regions, facilitating focused public health interventions, and optimising resource allocation. On the client-facing side, AI also promises to accelerate efforts to engage communities out of reach of existing efforts. For example, conversational agents powered by large language models can answer client questions about treatment and guide them to tailored care pathways using culturally sensitive and de-stigmatising language.7 AI tools using integrated computer vision and conversational agents are being scaled up across sub-Saharan Africa to support HIV self-testing, enabling real-time interpretation of results and reflexive linkage to appropriate services.8,9 These diverse applications show the potential of AI to not only effectively address the complex challenges that undermine efforts but also to offer a pathway towards more sustainable and client-centred HIV programming.
See article to read the full context
