MIT researchers show how to create “humble” AI
The MIT-led team is designing artificial intelligence systems for medical diagnosis that are more collaborative and forthcoming about uncertainty.
Massachusetts Institute of Technology
CAMBRIDGE, MA -- Artificial intelligence holds promise for helping doctors diagnose patients and personalize treatment options. However, an international group of scientists led by MIT cautions that AI systems, as currently designed, carry the risk of steering doctors in the wrong direction because they may overconfidently make incorrect decisions.
One way to prevent these mistakes is to program AI systems to be more “humble,” according to the researchers. Such systems would reveal when they are not confident in their diagnoses or recommendations and would encourage users to gather additional information when the diagnosis is uncertain.
“We’re now using AI as an oracle, but we can use AI as a coach. We could use AI as a true co-pilot. That would not only increase our ability to retrieve information but increase our agency to be able to connect the dots,” says Leo Anthony Celi, a senior research scientist at MIT’s Institute for Medical Engineering and Science, a physician at Beth Israel Deaconess Medical Center, and an associate professor at Harvard Medical School.
Celi and his colleagues have created a framework that they say can guide AI developers in designing systems that display curiosity and humility. This new approach could allow doctors and AI systems to work as partners, the researchers say, and help prevent AI from exerting too much influence over doctors’ decisions.
Celi is the senior author of the study, which appears today in BMJ Health and Care Informatics. The paper’s lead author is Sebastián Andrés Cajas Ordoñez, a researcher at MIT Critical Data, a global consortium led by the Laboratory for Computational Physiology within the MIT Institute for Medical Engineering and Science.
Instilling human values
Overconfident AI systems can lead to errors in medical settings, according to the MIT team. Previous studies have found that ICU physicians defer to AI systems that they perceive as reliable even when their own intuition goes against the AI suggestion. Physicians and patients alike are more likely to accept incorrect AI recommendations when they are perceived as authoritative.
In place of systems that offer overconfident but potentially incorrect advice, health care facilities should have access to AI systems that work more collaboratively with clinicians, the researchers say.
“We are trying to include humans in these human-AI systems, so that we are facilitating humans to collectively reflect and reimagine, instead of having isolated AI agents that do everything. We want humans to become more creative through the usage of AI,” Cajas Ordoñez says.
To create such a system, the consortium designed a framework that includes several computational modules that can be incorporated into existing AI systems. The first of these modules requires an AI model to evaluate its own certainty when making diagnostic predictions. Developed by consortium members Janan Arslan and Kurt Benke of the University of Melbourne, the Epistemic Virtue Score acts as a self-awareness check, ensuring the system’s confidence is appropriately tempered by the inherent uncertainty and complexity of each clinical scenario.
With that self-awareness in place, the model can tailor its response to the situation. If the system detects that its confidence exceeds what the available evidence supports, it can pause and flag the mismatch, requesting specific tests or history that would resolve the uncertainty, or recommending specialist consultation. The goal is an AI that not only provides answers but also signals when those answers should be treated with caution.
“It’s like having a co-pilot that would tell you that you need to seek a fresh pair of eyes to be able to understand this complex patient better,” Celi says.
Celi and his colleagues have previously developed large-scale databases that can be used to train AI systems, including the Medical Information Mart for Intensive Care (MIMIC) database from Beth Israel Deaconess Medical Center. His team is now working on implementing the new framework into AI systems based on MIMIC and introducing it to clinicians in the Beth Israel Lahey Health system.
This approach could also be implemented in AI systems that are used to analyze X-ray images or to determine the best treatment options for patients in the emergency room, among others, the researchers say.
Toward more inclusive AI
This study is part of a larger effort by Celi and his colleagues to create AI systems that are designed by and for the people who are ultimately going to be most impacted by these tools. Many AI models, such as MIMIC, are trained on publicly available data from the United States, which can lead to the introduction of biases toward a certain way of thinking about medical issues, and exclusion of others.
Bringing in more viewpoints is critical to overcoming these potential biases, says Celi, emphasizing that each member of the global consortium brings a distinct perspective to a broader, collective understanding.
Another problem with existing AI systems used for diagnostics is that they are usually trained on electronic health records, which weren’t originally intended for that purpose. This means that the data lack much of the context that would be useful in making diagnoses and treatment recommendations. Additionally, many patients never get included in those datasets because of lack of access, such as people who live in rural areas.
At data workshops hosted by MIT Critical Data, groups of data scientists, health care professionals, social scientists, patients, and others work together on designing new AI systems. Before beginning, everyone is prompted to think about whether the data they’re using captures all the drivers of whatever they aim to predict, ensuring they don’t inadvertently encode existing structural inequities into their models.
“We make them question the dataset. Are they confident about their training data and validation data? Do they think that there are patients that were excluded, unintentionally or intentionally, and how will that affect the model itself?” he says. “Of course, we cannot stop or even delay the development of AI, not just in health care, but in every sector. But, we must be more deliberate and thoughtful in how we do this.”
###
The research was funded by the Boston-Korea Innovative Research Project through the Korea Health Industry Development Institute.
Journal
BMJ Health & Care Informatics
Article Title
Engineering framework for curiosity-driven and humble AI in clinical decision support
Article Publication Date
23-Mar-2026
Virtual reality shown to improve medical students' understanding of head and neck anatomy
Pilot study finds VR-based learning boosts anatomical knowledge and confidence regardless of prior technology experience
American Academy of Otolaryngology - Head and Neck Surgery
A new study published in OTO Open, the open-access journal of the American Academy of Otolaryngology–Head and Neck Surgery Foundation (AAO-HNSF), finds that a standardized virtual reality (VR) educational experience improved medical students' knowledge and confidence in head and neck anatomy—and did so regardless of students' prior experience with VR or video gaming.
“The anatomy of the head and neck is one of the most spatially complex regions in medicine. Virtual reality gives learners the ability to step inside that anatomy and explore it in three dimensions in a way that textbooks and static images simply cannot. What’s also exciting is that these immersive learning tools can be accessible and beneficial for all medical trainees,” said corresponding author Michael Yim, MD, Otolaryngology Program Director and Associate Professor of Otolaryngology and Neurosurgery at LSU Health Shreveport.
This pilot study evaluated whether a commercially available VR platform could serve as an effective supplement to traditional cadaveric anatomy training. Twenty-one medical students, all of whom had previously completed a formal cadaveric head and neck anatomy course, participated in a guided, immersive VR session.
The VR platform received high ratings from participants for control, sensory immersion, and realism, with minimal distraction or frustration reported. Standardized assessments of workload and presence—the NASA Task Load Index and Presence Questionnaire—confirmed that students were able to engage effectively with the virtual environment with low stress and high perceived success, even those with no prior VR experience.
The authors note that this is the first study to evaluate a VR adjunct specifically for head and neck anatomy education and call for larger, multi-institutional studies and prospective trials comparing VR-based learning directly to conventional teaching methods.
Study Citation: Alvarez, I., Johnson, E., Latour, M. and Yim, M.T. (2026), Next Dimension Medical Education: A Pilot Study Exploring Virtual Reality in Head and Neck Anatomy. OTO Open, 10: e70217. https://doi.org/10.1002/oto2.70217
OTO Open is the official open-access journal of the American Academy of Otolaryngology--Head and Neck Surgery Foundation. Its mission is to publish clinically relevant, contemporary, and ethical research in otolaryngology--head and neck surgery that advances patient care and supports the global medical community through free and unrestricted access to peer-reviewed science.
About the AAO-HNS/F
The AAO-HNS/F is one of the world’s largest organizations representing specialists who treat the ears, nose, throat, and related structures of the head and neck. Otolaryngologist-head and neck surgeons diagnose and treat medical disorders that are among the most common affecting patients of all ages in the United States and around the world. Those medical conditions include chronic ear disease, hearing and balance disorders, hearing loss, sinusitis, snoring and sleep apnea, allergies, swallowing disorders, nosebleeds, hoarseness, dizziness, and tumors of the head and neck as well as aesthetic and reconstructive surgery and intricate micro-surgical procedures of the head and neck. The Academy has approximately 13,000 members. The AAO-HNS Foundation works to advance the art, science, and ethical practice of otolaryngology-head and neck surgery through education, research, and quality measurement.
Journal
OTO Open
Article Title
Next Dimension Medical Education: A Pilot Study Exploring Virtual Reality in Head and Neck Anatomy
