Case reports have been central to the development of clinical science from antiquity to the present. From the descriptive accounts of ancient Egyptian and Greek doctors to the more complex presentations of modern medicine, the case report has long served as an important genre for documenting novel or rare disorders and new treatments.¹

Although randomized clinical trials (RCTs) and large observational studies now dominate medical science, case reports have experienced a revitalisation in the era of online publishing.² The increasing number of online open-access journals has allowed for a substantial rise in published case reports.

Case reports remain valuable for many reasons.³ They frequently provide the first documentation of rare disorders, unexpected side effects, or emerging public health threats. Well-known examples include early reports of thalidomide-associated birth defects and initial clinical descriptions of COVID-19.^3,4 Increasingly, modern case reports also incorporate the patient’s voice, acknowledging the importance of the patient’s experience of illness and recovery.

Their pedagogical importance is equally notable.³ Case reports help students and early-career doctors develop their diagnostic reasoning skills and understand the complexities and uncertainties of clinical practice. The genre also represents an intuitive and accessible entry point into medical research and academic writing. Case reports do not require extensive funding, large cohorts, or multi-site collaboration, as is typically the case for more complex research designs such as RCTs.

However, the genre also has its challenges. Authors must distil extensive medical records into a brief and precise narrative without missing important information. And as is characteristic of qualitative methods, case reports have inherent limitations such as limited generalisability, an inability to infer causality, and a risk of over-interpretation.³

AI can be of help in case reporting. For instance, large language model-based tools (LLMs) can aid in synthesising large volumes of clinical documentation that may form the basis of a case report. LLMs can also help to identify relevant background literature, outline manuscripts, improve clarity of language, or aid in translation. As the technology advances rapidly, AI is likely to become increasingly central to scientific work, including in the preparation of case reports.

However, AI should be used with caution. There is a need to consider data privacy and relevant legislation in relation to the processing and storage of sensitive data. Relevant regulatory frameworks, such as the General Data Protection Regulation⁵ and the Health Insurance Portability and Accountability Act,⁶ place specific requirements on the handling of identifiable health information. Particular caution is needed when using cloud-based AI systems, where data may be processed in other jurisdictions. On-premise AI tools operating within secure environments may offer higher levels of data privacy. Moreover, AI models may hallucinate, i.e., fabricate information, and there may also be concerns relating to inadvertent plagiarism if text is sourced without proper referencing. Examples of appropriate AI use include translation or language editing, while using AI for clinical interpretation or drawing conclusions without verification would be inappropriate. Importantly, AI cannot replace a doctor’s clinical experience, understanding, or empathy.

Regardless of whether AI is used or not, the central steps of case reporting persist: selecting a case with educational or scientific value, reviewing the existing literature, obtaining patient consent and permissions—including for publication—in line with applicable data protection regulations, gathering the data and extracting the essential information, and writing up the report.

When AI tools are used, authors may consider the following points: (1) de-identify patient data before using AI, (2) comply with local policy and legislation, including relevant institutional requirements, (3) avoid uploading sensitive information to public AI tools, (4) verify all AI-generated outputs, (5) document prompts and AI contributions where possible, including basic audit trails and the AI model and version used, (6) disclose AI use, and (7) do not list AI tools as authors.

Reporting frameworks such as the CARE and SCARE guidelines provide valuable structure for case reports.^7,8 More recently, guidance specific to AI use, such as TITAN, describe steps that may help ensure a transparent, reproducible, and ethical research process.⁹ The use of AI is now also discussed in the recommendations by the ICMJE¹⁰ and COPE,¹¹ as well as in guidance from the World Association of Medical Editors¹² and major academic publishers.

Case reports have proven their value to medical science throughout the ages and are likely to stay relevant also in the age of AI.

In the preparation of this editorial, AI tools were used for language editing. All content was reviewed and approved by the author.