Half a century after Julian Tudor Hart described the inverse care law—that medical availability varies inversely with need—artificial intelligence may be turning this into a more dangerous, dynamic phenomenon, according to a new Lancet commentary. The piece argues that AI implementation is already unevenly distributed across the US healthcare system.
This clustering threatens to widen existing health inequities. The analysis examined 3,560 US hospitals from 2023–24 and found that regions with greater healthcare need were less likely to have hospitals using AI models. The pattern mirrors long-standing disparities in access to medical innovation.
The numbers tell a stark story: implementation was geographically concentrated, leaving vulnerable populations behind. The commentary builds on Hart's 1971 framework, which has shaped global health policy for decades. The authors warn that without deliberate intervention, AI could automate and accelerate inequity.
If left unchecked, this feedback loop could entrench disparities in diagnosis and treatment. The analysis calls for policy measures to ensure equitable distribution of AI tools. Hospital systems and regulators face pressure to address the geographic divide before it deepens.
The commentary does not specify which AI applications were studied or propose detailed solutions, limiting its immediate actionability. Further research is needed to confirm the pattern across different hospital types and regions.