Assessing the safety of patient-centred discharge medication instructions generated by an AI model
We found evidence of a difference in safety issues in AI-generated medication instructions for older, more complex patients, and patients with certain types of insurance. Health system and contextual differences could create unexpected variations in AI-generated outputs. Studies of AI-generated messaging for patients should consider the severity and likelihood of safety issues, localised trials, and ongoing auditing.
Authors
- Kristian Stanceski (ORCID: https://orcid.org/0000-0001-5165-4096)
- Michael Tang (ORCID: https://orcid.org/0000-0003-1936-6136)
- Sarita Lo (ORCID: https://orcid.org/0000-0002-5272-9958)
- Linda Ghali Koria (ORCID: https://orcid.org/0000-0002-6699-7080)
- Adam G. Dunn (ORCID: https://orcid.org/0000-0002-1720-8209)
- Shiyin Zhong (ORCID: https://orcid.org/0009-0007-2485-0332)
- K. Markova
- Julie Ayre
- Xumou Zhang
- Angela Pan
- Marguerite Tracy
- Jinman Kim
Institutions
- The University of Sydney (AU)
- Sydney Local Health District (AU)
Publication Details
- Journal
- International Journal of Medical Informatics
- Published
- 2026-01-08
- DOI
- https://doi.org/10.1016/j.ijmedinf.2026.106272
- Primary Topic
- Artificial Intelligence in Healthcare and Education
- Type
- article
- Field-Weighted Citation Impact
- 0.00