The World Health Organization has released guidance for the use of large language models and artificial intelligence in healthcare.
The guidance, released Jan. 18, proposes more than 40 recommendations for governments, technology companies and healthcare providers to ensure responsible use of large language models and AI.
The guidance outlines five applications for large language model use in health:
- Diagnosis and clinical care, including addressing written patient queries.
- Patient-guided utilization, for investigating symptoms and treatment.
- Clerical and administrative responsibilities, such as documenting and summarizing patient visits in EHRs.
- Medical and nursing education, involving simulated patient encounters for trainees.
- Scientific research and drug development, utilizing large language models to identify new compounds.
The guidance also outlines wider health system risks, including concerns about the accessibility and affordability of high-performing large language models.
WHO joins HHS and the Biden administration in releasing frameworks on how organizations can oversee AI in the healthcare sector.