The Fight Over Who Gets to Benefit From Medicine's Genomic Revolution

Your doctor can now read your entire genome for less than the cost of many everyday purchases. An AI system cleared by the FDA can spot a rib fracture your radiologist might miss. The science works. The question is whether your insurance will pay for it—and whether the algorithm was trained on bodies that look like yours.

Time magazine identifies rapid, low-cost genome sequencing, advances in precision medicine, and AI-driven diagnostic tools as breakthroughs reshaping U.S. health care. Yet whether these tools expand or restrict access depends on infrastructure, insurance coverage, and where you live—not just scientific progress.

What's Changed: Sequencing Costs Hundreds Instead of Millions, AI Spots Disease Faster Than Specialists

The cost of whole genome sequencing has fallen from $100 million in 2001 to $200–$400 per sample, with some high-throughput services offering sequencing for as low as $100. Commercial consumer services now offer clinical-grade whole genome sequencing with AI interpretation at prices ranging from $399 to $599.

AI diagnostic tools are proliferating rapidly, with the FDA clearing 295 AI/ML-enabled medical devices in 2025 alone, bringing the cumulative total to 1,451 authorized devices. Over 63% of those are designed specifically for diagnostic purposes, and radiology imaging accounts for 76% of all FDA-authorized AI medical devices. In February 2025, the FDA granted its first clearance of a foundation model-powered clinical AI device—Aidoc CARE1™ for rib fracture triage—marking a landmark in the field.

AI-driven precision medicine is cost-effective in 89% of evaluated cases, with precision oncology showing an incremental cost-effectiveness ratio of approximately $58,500 per quality-adjusted life year for high-risk cancer patients. Pharmacogenetic-guided treatments result in substantial savings, with 67% of patients seeing overall reductions in hospitalizations and pharmacotherapy expenses.

Early Adopters—But Not Yet Standard Care for Most Americans

Academic medical centers like Mayo Clinic and Geisinger are leading clinical implementation of genomic medicine. Some clinical laboratories now cap out-of-pocket charges or offer reduced self-pay options, with pharmacogenomic tests available for about $330 for patients whose insurance denies or limits coverage.

Insurance coverage for genome sequencing varies widely, heavily dependent on medical-necessity criteria, prior authorization requirements, and individual plan design. Medicaid coverage has expanded significantly, rising from 28% to 66% of Medicaid lives between 2024 and 2025, now available in 38 states for outpatient testing—though access remains uneven in rural and underserved communities. Approximately 18.3% of pediatric genetic testing requests face insurance denial, with private insurers denying tests at more than double the rate of public programs like Medicaid and Medicare.

The Infrastructure Barriers—Rural Gaps, Insurance Denials, Missing Datasets

Only 56% of rural and critical-access hospitals use predictive AI compared with 81% of urban health systems. The National Rural Health Association published a policy roadmap calling for broadband expansion and technical assistance to ensure AI closes rather than exacerbates rural access gaps.

Nearly 43% of patients recommended for genetic testing encounter insurance-related hurdles, and about half of eligible patients are denied access. Only 25% of appeals for denied genetic testing succeed, leaving many patients without access to potentially life-changing diagnostic information.

Dr. Richard Gibson of Mayo Clinic explains that many insurance policies were written before the genomic revolution, with definitions of "experimental" designed for traditional drug trials and procedures, not diagnostic technologies that have rapidly become standard of care. Dr. Marc Williams advises that when appealing denials, it's important to cite the most current version of relevant guidelines, including specific sections that support testing in particular clinical scenarios. Dr. Robert Green recommends including information about the costs of continued diagnostic uncertainty—additional specialist visits, hospitalizations, procedures, and ineffective treatments that could be avoided with a definitive diagnosis—when appealing genetic testing denials.

AI Trained Mostly on White, Wealthy Populations Can Misdiagnose Others

Genomic data remains overwhelmingly biased toward European ancestry, with 94.48% of genome-wide association studies focusing on European populations while African representation stands at only 0.47%. The proportion of non-European participants actually decreased from 19% in 2016 to 14% in 2021, indicating stagnation despite calls for more inclusive research.

Dr. Josh Denny of the NIH All of Us Research Program states that until now, over 90% of participants from large genomics studies have been of European descent, and the lack of diversity in research has hindered scientific discovery.

Polygenic risk scores and diagnostic tools derived from European cohorts show diminished accuracy when applied to non-European populations, leading to higher rates of variants of unknown significance in African, Asian, and Latin American groups. The median effect size of polygenic risk scores in African ancestry samples is only 42% of that in matched European samples. A polygenic risk score for body mass index explained 17.6% of variation in Europeans but only 2.2% in rural Ugandans, illustrating the dramatic performance gap.

A systematic review found that AI and machine learning systems in health care frequently underdiagnose conditions, mispredict mental health outcomes, and contribute to longer wait times for underrepresented populations, especially when training data overrepresents lighter-skinned and higher-income groups. Fay Cobb Payton, a health equity researcher at Rutgers-Newark, warns that preliminary research shows algorithms may be racially biased, even when Black patients are sicker than the rest of the population, leading to misdiagnosis, inadequate access to resources, or delays in treatment.

What Closing the Gap Would Require—And Why It Won't Happen Automatically

The NIH All of Us Research Program represents a notable exception, with roughly half of its data coming from individuals of non-European descent. The program has identified 275 million genetic variants absent from prior databases, demonstrating the scientific value of diverse genomic datasets. Dr. Josh Denny emphasizes that All of Us participants are leading the way toward more equitable representation in medical research through their involvement.

The Encoding Equity alliance was formed in 2024 by the Council of Medical Specialty Societies and the Doris Duke Foundation to identify incorrect uses of race in clinical algorithms and design accurate and equitable decision tools. Experts recommend promoting health and health care equity in all the different phases of an algorithm's life cycle, from data collection through deployment.

The World Health Assembly adopted a global resolution on precision medicine in May 2026 that explicitly addresses widening inequities in access to genomics, diagnostics, data science, and digital health. The resolution stresses underrepresentation of many populations—especially in low- and middle-income countries—in precision medicine research and urges Member States to build capacity so advances reduce, rather than exacerbate, health disparities.

The Stakes: These Technologies Can Either Reduce Disparities or Lock Them In Permanently

Genome sequencing and AI diagnostics represent a fork in the road for American health equity. The rapid pace of adoption—295 AI devices cleared in a single year, Medicaid coverage jumping from 28% to 66% in one year—means the window for shaping equitable access is narrow.

As Nature journal commentary argues, precision medicine without equity is just stratified inequality. Closing the gap requires policy mandates for diverse datasets, expanded Medicaid and Medicare coverage for genomic testing, investment in rural broadband and community health infrastructure, and regulatory oversight of AI bias. The choices about who benefits are being made right now—in insurance company coverage policies, FDA regulatory frameworks, NIH research priorities, and state Medicaid decisions.

Whether these tools democratize American health care depends not on the technology itself, but on whether policymakers, insurers, and health systems choose equity as an explicit goal.