Recent evidence suggests LLMs and ChatGPT can be useful supplements for cancer patient education but are not yet reliable enough to replace clinicians; they improve access to information and can help patients understand complex concepts, yet they still produce errors, hallucinations, and overly technical or biased content, especially in complex treatment decisions.
Several studies from 2023–2025 show that AI chatbots powered by LLMs can provide accurate basic information, personalized educational content, and 24/7 support, leading to improved patient recall and satisfaction, but also reveal risks of misinformation, misaligned treatment recommendations, and privacy concerns.
Oncology bodies and researchers increasingly recommend that LLMs be used only as clinician‑supervised, adjunct tools—pre‑vetted, curated, and framed transparently to patients—to balance educational benefits against the potential for harm.
Sources:
Errors and half-truths plague cancer treatment plans generated by ...
Evidence-Based Analysis of AI Chatbots in Oncology Patient ... - PMC
LLMs in oncology: ESMO's framework for integration in the clinic
Exploring Capabilities of Large Language Models such as ChatGPT ...
Can AI Chatbots Correctly Answer Questions about Cancer? - NCI
Application of AI Chatbot in Responding to Asynchronous Text ...
Doctors Say AI-Created Cancer Education Materials Easier to Read
ChatGPT shows limited ability to recommend guidelines ... - ecancer
Artificial Intelligence Chatbot as a Companion for Cancer Patients ...
The use of large language models to enhance cancer clinical ...
AI Chatbot Enhances Cancer Care - UHN Research
Assessing ChatGPT's Educational Potential in Lung Cancer ...
Medical accuracy of artificial intelligence chatbots in oncology
Review of Large Language Models for Patient and Caregiver ...