Large language models: rheumatologists’ newest colleagues?

In 2023, large language models demonstrated potential for use in rheumatology to accurately suggest diagnoses and provide empathetic patient education. However, the propensity of this technology to generate misleading information continues to pose risks. Balancing innovation with physician guidance is essential.

Key advances

• Large language models (LLMs) have demonstrated diagnostic potential for rheumatic diseases, in some situations even outperforming diagnosis by rheumatologists⁵.

• ChatGPT and other LLMs can potentially contribute to patient education by providing detailed and empathetic responses to their queries, thereby enhancing patient understanding, fostering trust and promoting treatment adherence⁶.

• Caution should be displayed in the use of LLMs, as they have displayed a propensity for generating ‘hallucinated’ or misleading information about anchor medications such as methotrexate⁷.