Conclusions

This pilot study provides the first experimental evidence that generative artificial intelligence, when appropriately configured with domain-specific instructions, can simulate divergent medico-legal reasoning in real cases of alleged healthcare liability. These findings should be interpreted with caution given the exploratory design and limited sample size. The ability of the customized GPT models to generate coherent, structured and perspective-aligned medico-legal reports within seconds underscores the potential value of AI as an assistive tool in the early analytical stages of medico-legal practice.

The differential performance observed between the patient-oriented and hospital-oriented models highlights the extent to which AI-driven outputs are influenced by role conditioning, reinforcing the need for carefully designed governance frameworks and strict oversight by qualified experts. While the models demonstrated strong capabilities in clinical synthesis and produced impairment estimates that were often concordant with human evaluations, the presence of hallucinated or formally inaccurate citations confirms that AI cannot independently ensure evidentiary reliability.

Taken together, these findings suggest that generative AI should be considered a promising adjunct for organizing documentation, exploring alternative argumentative perspectives, and supporting preliminary medico-legal analyses, but not as a substitute for human expertise. The medico-legal domain requires interpretative nuance, contextual understanding, and ethical accountability that remain beyond the reach of current AI systems.

Future work should expand the evidence base by incorporating larger and more diverse case series, integrating jurisdiction-specific medico-legal frameworks, and developing safeguards that mitigate the risks associated with automated citation generation and interpretative errors. As generative AI continues to evolve, interdisciplinary collaboration among forensic physicians, data scientists, and legal scholars will be essential to ensure that technological innovation enhances the rigor, transparency, and fairness of medico-legal assessments.