Post written by Roberta Maselli, MD, PhD, from the Department of Biomedical Sciences, Humanitas University, Pieve Emanuele, Italy.

This study evaluates the effectiveness and reliability of Chat Generative Pre-Trained Transformer 4.0 (Chat-GPT 4.0; OpenAI, San Francisco, Calif, USA) in addressing common patient queries regarding GI endoscopy. By assessing responses in terms of reliability, accuracy, and comprehensibility, the study explores the potential role of artificial intelligence (AI)—based conversational tools in patient education and preparation for endoscopic procedures.

Patient education is a key component of successful GI endoscopic procedures, but many individuals remain inadequately informed about aspects such as preparation, dietary restrictions, sedation, and potential risks. Chat-GPT 4.0, as a widely accessible AI tool, has the potential to enhance patient understanding and compliance. However, its effectiveness and limitations in delivering medical information need to be critically assessed before its integration into clinical practice.

This study found that Chat-GPT 4.0 provided highly reliable and accurate responses to most patient queries, particularly on general topics such as dietary guidelines, procedural expectations, and post-endoscopy care. Yet, certain areas—such as bowel preparation, medication management, and pacemaker-related concerns—showed reduced reliability because of the AI’s tendency to include regionally unavailable prep solutions or provide vague responses.

These findings suggest that AI-based patient education tools could complement clinical practice but require refinements, such as training with region-specific guidelines and expert validation. Future research should focus on integrating AI-driven responses with real-time physician oversight and evaluating patient comprehension and satisfaction.

Although Chat-GPT 4.0 demonstrates promise in assisting patient education, its responses must be used cautiously and supplemented with direct physician input for complex or regionally specific medical queries. Developing AI models incorporating up-to-date medical guidelines and real-world clinical validation will be essential for optimizing their role in healthcare settings.

Mean values of reliability, accuracy, and comprehensibility of 18 questions included in the study.

Visit iGIE’s Facebook, X/Twitter, LinkedIn, Instagram, and YouTube accounts for more content from the ASGE peer-reviewed journal that launched in December 2022.

Read the full article online.

The information presented in Endoscopedia reflects the opinions of the authors and does not represent the position of the American Society for Gastrointestinal Endoscopy (ASGE). ASGE expressly disclaims any warranties or guarantees, expressed or implied, and is not liable for damages of any kind in connection with the material, information, or procedures set forth.