Related Experiment Videos
Evaluation of an Artificial Intelligence Communication Platform for Racial Biases in Pretest Education About Prostate
Johanna Balas1, Lauren Bowling2, Nonna Shakhnazaryan3
1Department of Medicine, University of California San Francisco, San Francisco, CA.
JCO Clinical Cancer Informatics
|July 20, 2026
Summary
This study evaluated an AI chatbot for prostate cancer genetic testing education. The AI tool showed no racial bias in its responses, suggesting potential for equitable genetic education.
Area of Science:
- Oncology
- Medical Informatics
- Health Equity
Background:
- Germline testing in advanced prostate cancer (PCa) is crucial for treatment decisions and risk assessment.
- Black patients are less likely to complete germline testing compared to White patients.
- Artificial intelligence (AI) tools are emerging for pretest education but require evaluation for racial bias.
Purpose of the Study:
- To develop and evaluate ProGene, a generative AI chatbot for prostate cancer germline testing education, specifically for Black patients.
- To assess ProGene's responses for racial bias across comprehensiveness, accuracy, readability, actionability, and quality.
Main Methods:
- ProGene, an AI chatbot, was prompted with questions about PCa germline testing using three vignettes (Black, White, race-agnostic).
- Responses were evaluated by blinded reviewers across five domains: comprehensiveness, accuracy, readability (SMOG, Flesch-Kincaid), actionability, and quality (DISCERN-AI).
- Outcomes were compared between racial groups and question types using statistical tests.
Main Results:
- No statistically significant racial disparities were found in ProGene's responses across the five evaluation domains.
- Mean comprehensiveness was 67%, with lower scores for Black vignettes on specific questions (testing types).
- Inaccuracies (32%) occurred mainly in sample collection and cost/insurance, without racial variation; readability, actionability, and quality were generally good.
Conclusions:
- The AI chatbot ProGene demonstrated no statistically significant racial bias in educating patients about prostate cancer germline testing.
- Responses were generally of good quality and actionable, indicating AI's potential to promote equity in genetic education.
- Deficiencies in comprehensiveness and accuracy necessitate refinement and/or human oversight for effective AI implementation in genetic counseling.
Related Concept Videos
02:16Surveys
Often, psychologists develop surveys as a means of gathering data. Surveys are lists of questions to be answered by research participants, and can be delivered as paper-and-pencil questionnaires, administered electronically, or conducted verbally. Generally, the survey itself can be completed in a short time, and the ease of administering a survey makes it easy to collect data from a large number of people.
01:27Halo Effect
The halo effect is a cognitive bias in which an individual's overall impression influences judgments about their specific traits. This psychological phenomenon leads people to associate positive characteristics with those they perceive as generally good and negative characteristics with those they view as bad. This effect is particularly influential in social perception, professional evaluations, and decision-making processes.The Psychological Basis of the Halo EffectThe halo effect is rooted...
02:55Stereotypes, Prejudice, and Discrimination
Humans are very diverse and although we share many similarities, we also have many differences. The social groups we belong to help form our identities (Tajfel, 1974). These differences may be difficult for some people to reconcile, which may lead to prejudice toward people who are different. Prejudice is a negative attitude and feeling toward an individual based solely on one’s membership in a particular social group (Allport, 1954; Brown, 2010). Prejudice is common against people who are...