AI and Breast Cancer Disparities: Can Technology Help Close the Gap?

How artificial intelligence is poised to transform breast cancer care across different populations

From standard X-rays in the early 20th century, to the invention of the mammogram in 1930, all the way to 3D mammography and other advanced tools today, the history of breast cancer screening has been one of profound innovation. The latest breakthroughs incorporate artificial intelligence (AI) to further classify risk. AI breast cancer risk screening is now available nationwide, and the National Comprehensive Cancer Network (NCCN) even included it in its 2026 guidelines.

But whether or not these tools will work for—and reach—every woman remains to be seen. The Breast Cancer Research Foundation (BCRF) is committed to making these game-changing technologies accessible to women everywhere. Ahead, find out how AI plays a role in addressing breast cancer disparities.

AI breast cancer risk testing is now available nationwide

AI is revolutionizing many facets of medicine, including how experts determine breast cancer risk. As of October 2026, Clairity Breast—the first U.S. Food & Drug Administration (FDA)-authorized AI tool that predicts five-year breast cancer risk from a routine mammogram—is available nationwide through Everlywell, a digital health and diagnostic testing company. Doctors can recommend people with a qualifying mammogram from the past 12 months for analysis of their screening using this first-of-its-kind tool, which was made possible with BCRF support.

AI software learns patterns from a large number of mammograms. In addition to helping doctors estimate future breast cancer risk, it can also help radiologists read scans. Experts are seeing the potential: In April 2026, the NCCN updated its breast screening guidelines to recommend AI-based risk assessment starting at age 35. Those with a five-year risk of 1.7% or higher are considered at increased risk.

Why older risk tools missed some women

Established breast cancer risk models like Gail and Tyrer-Cuzick rely on family history and genetics to make calculations. But only a small number of breast cancers—about 5-10%—are hereditary. That means that up to 95% of patients have no significant family history or known mutation. These tools were also built largely on data from women of European ancestry, so they have not generalized well for women of color.

Where AI could help close the gap

Tools such as Clairity Breast can determine a person’s risk of breast cancer by interpreting the mammogram itself. Instead of relying on family history, it looks for patterns in mammograms that indicate risk. Clairity Breast was also developed with proportional representation of racial and ethnic groups, ensuring that women of different backgrounds can accurately assess their breast cancer risk. This is especially important considering that Black women ages 20-39 are more likely to develop breast cancer compared to younger women of other ethnicities.

Where AI could widen the gap

Using AI tools to interpret breast cancer risk is a powerful step forward, but it does come with some potential risks. One example is bias: When the data used to train an AI tool doesn’t reflect a wide range of participants, it may work better for some groups than others. That’s why it’s so important to have a representative pool of data from which these technologies draw. A 2024 Duke study found that an FDA-approved algorithm was more likely to give a false-positive result to Black women, and less likely to Asian women, compared to white women. False positives lead to extra workups and worry for patients. Another review from the European Journal of Cancer found that out of 264 AI mammography studies, just 0 to 25% reported patients’ race or ethnicity (with most patients identifying as white), and nearly all came from high-income countries. No studies from low-income countries were used.

While AI has the potential to make care more accessible, technologies remain limited, especially in low-income and rural settings. Cost and insurance coverage also may limit who has access to these new tools.

To move forward fairly, AI tools need diverse data, transparency about who they were tested on, and underrepresented groups involved in the research.

Questions to ask about AI at your next mammogram

Here are some questions to ask at your next breast cancer screening:

• Is AI used to read my mammogram or estimate my risk?
• Was this tool tested on someone like me?
• What happens if AI flags something or gives me a higher risk score?
• Should I have a breast cancer risk assessment before 40? (The American College of Radiology recommends one by age 25.)

How BCRF is helping AI work for everyone

BCRF is working to ensure all women can benefit from AI-supported breast cancer screening tools. BCRF supported Clairity Breast’s testing, and, in 2025, announced a second round of funding. Other BCRF-funded researchers are focused on developing strategies to integrate AI into routine screening in India, and deploying AI in digital pathology research in the UK.

The AI tools of the future will be in no small part influenced by the choices made by researchers, developers, and funders today. For example, BCRF’s Health Equity Initiative investigators—comprised of researchers who are working to reduce breast cancer disparities in outcomes for Black women—has assembled a database of Black women with breast cancer. They have compiled genetic, tumor, and social drivers of health data from more than 5,000 participants, the largest database of its kind. This is the kind of data that future AI tools will need to learn from to produce accurate findings, particularly in Black women.

Help fuel more breakthroughs by donating to BCRF here.

Frequently asked questions