The Role of Artificial Intelligence (AI) in Cancer Care
- AI can empower patients, helping them decode medical jargon, understand complex treatments, and arrive at appointments better prepared for conversations with their oncologists, which is a trend Dr. Alan Kerr at Tampa General Cancer Institute says can boost patient confidence and engagement.
- It’s also important to understand AI’s limitations, Dr. Kerr says.
- AI can be misleading, especially when interpreting scans or lab results. Nearly half of chatbot-generated medical answers in a 2026 BMJ Open study were “problematic” in some way (misleading, not properly cited, etc.).
- Dr. Kerr stresses that AI can be used as a support tool, not a substitute for guidance from medical professionals and specialists.
- Medically-guided AI platforms like SurvivorNet’s My Health Questions can also provide patients with safer, doctor‑supported answers.
Dr. Alan Kerr, a malignant hematologist at Tampa General Cancer Institute, says the tools can be helpful for patients, but they should also be wary of what AI is telling them.
Read MorePros & Cons of AI in Cancer Care
“Some people will know the difference in the mechanism of action between [leukemia treatments] Brukinsa and Jaypirca because they’ve already looked it up coming in,” Dr. Kerr explains. “Then they have the question for me: which one do we start first? It’s great to have these tools for patients so that they can educate themselves.”He believes this kind of engagement can be genuinely beneficial. Patients who understand their disease and treatment options often feel more confident and more in control. However, AI should not be mistaken as fool-proof when it comes to health and medicine.
The challenge, Dr. Kerr says, is that AI is not perfect. It can misunderstand medical images, misinterpret lab results, or present outdated or oversimplified information.
“You have to be careful with it,” he warns. “I have had patients use AI and try to interpret their PET scans and things, and they have shown up in my clinic in emotional distress because AI gave them an inaccurate answer.”
He cautions that one of AI’s biggest risks is that it can sound confident even when it’s wrong.
How to Fact-Check Misleading Information
AI models can only rely on the information fed into them. If the data is incomplete, inaccurate, or not tailored to a patient’s specific medical situation, the output can be misleading.
A 2026 study in the medical journal BMJ Open suggests that many mainstream chatbots still aren’t ready for medical decision‑making. Researchers found that nearly half of AI‑generated answers to health questions were “problematic,” often due to misinformation, missing context, or overly confident but incorrect explanations.
Researchers evaluated several major generative AI systems — Google’s Gemini, High‑Flyer’s DeepSeek, Meta AI, OpenAI’s ChatGPT, and XAI’s Grok — to see how well they handled medical information, citations, and readability.
Each chatbot was asked 10 questions across cancer, vaccines, stem cells, nutrition, and athletic performance.
Key findings include:
- 49.6% of responses were problematic, including 19.6% that were highly problematic and potentially harmful if taken literally.
- No chatbot clearly outperformed the others, though Grok produced more highly problematic answers than expected.
- Answers about vaccines and cancer were generally stronger, while responses about stem cells, athletic performance, and nutrition were the weakest.
- Many answers were written at an advanced level, making them hard for the average patient to understand.
- Citation quality was subpar across the board, with some incomplete references.
The key takeaway of the research is that general‑purpose chatbots are not yet reliable for medical guidance, especially in areas where misinformation is common or evidence is evolving.
Expert Resources for CLL Patients
- Planning Your CLL Treatment Journey: How Doctors Choose Your First Therapy
- A Decade of Progress: New Targeted Treatments Are Redefining What’s Possible For People With CLL
- The Workup For CLL: How Doctors Diagnose This Type of Blood Cancer
- Understanding BTK Inhibitors: Uses and Limitations in CLL Treatment
“There is a lot of misinformation online, and AI decision‑making models are not perfect,” Dr. Kerr says. “It all depends on the information that is fed into that particular algorithm.”
This is why oncologists emphasize that AI should be used to “support” your care, not replace professional medical interpretation.
Dr. Kerr encourages patients to keep learning, keep asking questions, and keep using tools that help them feel informed. But he also stresses one essential step: include your care team.
“For those patients who are out there looking up their own information, just make sure you take your questions to your oncologist. Say, ‘Hey, here’s what I looked up.’ I wouldn’t get too upset if you look something up and it gives you an answer that scares you. I would just make sure you talk to your oncologist first, and they can help you interpret that information.”
A Safer Way to Use AI: Tools Built With Doctors
One example of a more reliable approach is SurvivorNet’s doctor‑supported AI tool, My Health Questions.
Unlike general‑purpose AI platforms, this tool is specifically designed for cancer patients and grounded in medical expertise. SurvivorNet’s reporting highlights how patients are using it to get clearer, more trustworthy answers to critical questions about their diagnosis, treatment options, and side effects.
Since “My Health Questions” is built with physician oversight and tailored to oncology topics, it helps reduce the risk of misinformation and gives patients a safer way to prepare for conversations with their care team.
AI can help patients feel informed, organized, and empowered. It can make complex cancer information more understandable and accessible. However, it can also misinterpret, oversimplify, or inadvertently frighten patients, especially when used to analyze personal medical data.
Learn more about SurvivorNet's rigorous medical review process.
