Can DALL-E 3 reliably generate 12-lead ECGs and teaching illustrations? This study explores the feasibility of using DALL-E 3 to generate medical illustrations, focusing on 12-lead ECGs and CPR teaching materials. DALL-E 3, integrated with ChatGPT, can generate images based on text prompts. The study found that DALL-E 3 can create basic 12-lead ECGs with some elements, but the waveforms are not physiologically accurate. The heart rate and T waves were not correctly represented, and the ECGs lacked standard features. However, DALL-E 3 produced accurate CPR illustrations with proper hand placement and technique. For ECG teaching, DALL-E 3 created heart-shaped waveforms that were creative and engaging, though not entirely accurate. The study concludes that while DALL-E 3 shows potential for generating basic medical illustrations, further training and expert validation are needed for reliable clinical use. The CPR illustrations were more promising, with clear and accurate representations of techniques. However, some spelling errors were present in the illustrations. The ECG teaching illustration, while creative, lacked accurate waveform details. Overall, DALL-E 3 has potential for simple medical illustrations but requires refinement and expert validation for complex, clinically accurate outputs. The study suggests that further development and training data could enhance the accuracy and reliability of DALL-E 3 in medical education and clinical settings.Can DALL-E 3 reliably generate 12-lead ECGs and teaching illustrations? This study explores the feasibility of using DALL-E 3 to generate medical illustrations, focusing on 12-lead ECGs and CPR teaching materials. DALL-E 3, integrated with ChatGPT, can generate images based on text prompts. The study found that DALL-E 3 can create basic 12-lead ECGs with some elements, but the waveforms are not physiologically accurate. The heart rate and T waves were not correctly represented, and the ECGs lacked standard features. However, DALL-E 3 produced accurate CPR illustrations with proper hand placement and technique. For ECG teaching, DALL-E 3 created heart-shaped waveforms that were creative and engaging, though not entirely accurate. The study concludes that while DALL-E 3 shows potential for generating basic medical illustrations, further training and expert validation are needed for reliable clinical use. The CPR illustrations were more promising, with clear and accurate representations of techniques. However, some spelling errors were present in the illustrations. The ECG teaching illustration, while creative, lacked accurate waveform details. Overall, DALL-E 3 has potential for simple medical illustrations but requires refinement and expert validation for complex, clinically accurate outputs. The study suggests that further development and training data could enhance the accuracy and reliability of DALL-E 3 in medical education and clinical settings.