Performance of GPT-5.2 in handling pharmaceutical calculations and interpreting medication prescriptions: Implications in pharmacy education and clinical practice

Pharmaceutical calculations and prescription interpretations are two integral components of medical education and responsibilities of healthcare providers, especially pharmacists. Various studies have highlighted the potentials of artificial intelligence-powered chatbots, namely ChatGPT (OpenAI), in various fields of medical and healthcare sciences. Herein, we provide a qualitative assessment of the ability of GPT-5.2 in performing pharmaceutical calculations and interpreting prescriptions. A total of 20 calculation and 5 prescription questions were developed by the authors and used for GPT-5.2 evaluation. Then, GPT-5.2 was assessed on providing feedback and utilized sources. As GPT-5.2 was asked to provide the final answer and then explain the calculation steps, it initially answered 17 out of the 20 calculation questions correctly (85%). When detailing the steps it undertook to find the solutions; however, GPT-5.2 was able to correct itself and thereby answer all questions correctly. In addition, it explained the approach it took, detailed the calculation steps and remained precise to the questions’ points. The responses were rapid, concise, accurate, clear and explanatory in nature. Regarding prescription questions, it answered all of them correctly and clarified the meanings of each abbreviation alone. When asked to provide feedback, GPT-5.2 was able to “remember” the number of prompted calculation questions, while providing an accurate rating of their difficulty and list of utilized resources. In conclusion, GPT-5.2 can handle pharmaceutical calculations and interpret prescriptions and medication orders, but is still prone to errors, with significant implications in pharmacy education and clinical practice.