The development of large language models such as ChatGPT and, more broadly, generative artificial intelligence (gen AI) has undoubtedly been remarkable. These AI models, which can learn patterns from existing data to create new, synthetic text, images and video, have forced us to fundamentally rethink how we work — or what work means. What is also Health care is no exception. As resident physicians, we are excited by reports of increasingly general AI use cases in healthcare systems, from helping doctors respond to patient messages and ordering diagnostic tests to tracking gaps in patient care. From doing and designing “smart hospital rooms” that integrate sensors to protect patients. Monitoring
However, the superiority of these initiatives is emerging in a few institutions. As we continue to develop and experiment with AI technologies, equitable access to these technologies is critical to preventing a growing disparity in healthcare quality. We must work to ensure that smaller clinics, community hospitals, and underfunded institutions are not left behind.
Drivers of AI Disparity
Innovative uses of AI in higher institutions are admirable and exciting. But in most other hospitals that don't have the resources of elite research universities, adoption of, or even awareness of, general AI tools is lacking or absent.
At our institution, for example, there was probably one teaching session (over 120) during this past academic year that specifically discussed AI, and only then at a broad, superficial level. Opportunities for trainees to learn more about gen AI tools are limited due to lack of awareness and access. Our experiences likely represent the overwhelming norm of residents and clinicians across the country.
This disparity is influenced by several factors. First, there are financial constraints. Most institutions do not have funds and funds for research and development of AI technologies. In a self-sustaining cycle, underfunded institutions also face challenges in obtaining funding due to poor grant-writing skills and connections to attract investment.
Then there are educational gaps. The expertise required to effectively implement gen AI and oversee its use is significant, often requiring the creation of entire teams or offices (for example, Stanford has a dedicated “Artificial Intelligence Clinical Integration” is the team). Hospitals that have been successful in deploying AI often employ the skills of engineers and researchers drawn from affiliated engineering schools — the affiliation is non-existent in most hospitals. This educational gap extends to medical schools, where the curriculum often does not adequately cover the latest advances in AI.
Third, at elite research institutions, driving innovation is often strongly supported by hospital management, expressed through investments in AI programs and strategic partnerships. Smaller hospitals may struggle with managerial inertia, where leadership is either unaware of AI's potential or reluctant to invest in new technologies due to uncertainty about the return on investment. Finally, cultural resistance to AI still exists. Healthcare professionals may be concerned about the possibility that gen AI could fundamentally alter their role. This reluctance is often rooted in a lack of understanding of the ways that AI can complement rather than replace human expertise.
The Implications of the Creative AI Disparity
Scientific and medical progress has always been brought about by a select few institutions. So why is disparity important in general AI implementation? We believe the rapid pace of AI development and the sheer magnitude of its impact warrants special consideration.
Quality of patient care
AI promises to enhance patient care by improving diagnostic accuracy and patient monitoring, creating more personalized treatment plans, and providing real-time clinical decision support — all faster. On speed. By automating documentation and other non-patient-facing tasks, AI indirectly allows clinicians to spend more time helping patients feel cared for.
If these technologies live up to their promises, disparities in AI access and capability will lead to variations in patient care. In the long term, populations served in AI-equipped hospitals will likely experience lower rates of chronic diseases, longer life expectancies, and better overall health outcomes.
Economic efficiency
One of the most common applications of AI is to improve operational efficiency in healthcare delivery. This includes streamlining administrative tasks, clinical workflow, and resource management. Hospitals can transfer the resulting economic savings to further improve patient care. However, ironically, well-resourced hospitals are best positioned to realize these benefits. Small and underfunded facilities are unable to leverage AI for administrative tasks and must continue to rely on laborious and expensive processes.
Technical lockout
The pace of AI development raises the risk of “technological lockout,” in which organizations that lag behind in AI adoption find it increasingly difficult to catch up.
As well-funded institutions continue to invest in the latest technologies and modify elements of their health systems (such as electronic medical records) to incorporate these tools, they can adapt to their specific needs and new ones. Will develop proprietary AI solutions that adapt to the challenges. Maintaining your competitive edge. Additionally, gen AI thrives on large amounts of high-quality data. Leading institutions often have access to comprehensive datasets and infrastructure to effectively manage and analyze this data, while smaller hospitals cannot keep up, limiting their ability to optimize AI models and resulting in significant technological gaps over time.
As certain institutions demonstrate the ability to innovate in gen AI, they will also attract top talent skilled in these tools, creating a concentration of expertise in well-funded hospitals. Healthcare workers in small or underfunded hospitals also have limited opportunities for training in AI technologies, perpetuating a skills gap. Finally, well-resourced hospitals are often more adept at influencing and navigating policy and regulatory frameworks, leveraging their technical capabilities to ensure compliance.
Hospitals ill-equipped to do so will face delays in implementing AI solutions. Over time, a fragmented healthcare landscape may result, with many organizations constantly advancing technologically while others remain stagnant.
Bridging the gap
Addressing the widening disparity in AI will require broad strategies that include government intervention, educational initiatives, collaborative models, and support from both the public and private sectors.
Government and Policy Intervention
Government policies can play an important role in promoting fair AI implementation. Policies should focus on providing funding, training grants, and partnership mandates that encourage AI adoption in small, underfunded, and community hospitals. There are precedents for this. Past initiatives have supported the adoption of electronic health records and health IT. Regulations should ensure that AI technologies address local health challenges and are distributed equitably across regions.
Education and training programs
To reduce the educational gap, initiatives to increase general AI knowledge at all levels of medical education are necessary. Professional associations such as the Association of American Medical Colleges have developed resources for this purpose and should continue to offer guidance on the design of medical school curricula and professional development programs. Healthcare systems can partner with academic and corporate organizations to create institution-specific AI training modules, as evidenced by the abundance of existing online courses on the use of general AI.
Collaborative models
Building collaborative models for resource allocation between AI-equipped hospitals and other hospitals has vast potential to reduce disparities. Well-resourced and innovation-focused hospitals should provide guidance and technical support to underfunded or smaller hospitals. Establishing regional AI “hubs” that serve as centers of excellence can facilitate the sharing of knowledge and resources. Meanwhile, less AI-equipped hospitals should actively consider how general AI can benefit their workflows.
Similarly, the private sector should be encouraged to invest in affordable AI solutions that fit the needs of underfunded and community hospitals. By developing small, underfunded hospitals, the entire system becomes more flexible and able to deal with both public health crises and everyday medical problems alike.
Given how fast AI tools are evolving, it's not premature to continue developing gen AI in an equal way. Failure to do so risks creating gaps that will be even more difficult to bridge. By implementing the strategies described above, we can meet our ethical imperative to achieve a more inclusive healthcare system in which AI technologies benefit all patients, regardless of where they are or where they are. Resources available to their health care providers.
Henry Bair, MD, MBA, and Mak Djulbegovic, MD, MSc, are resident physicians at Wills Eye Hospital/Jefferson Health in Philadelphia. Baer previously directed several courses on digital health at Stanford University School of Medicine. Djulbegovic is an AI researcher whose work focuses on biomedical applications of large language models.