What Healthcare’s AI Readiness Gap Reveals About The Next Phase Of Adoption
Mark Thomas, Chief Technology Officer, MRO.
gettyAI is becoming more embedded in healthcare every day. As health systems explore ways to apply AI across a growing range of functions, the right foundation becomes more important.
This is especially critical in healthcare quality. Quality operations rely on clinical context, accurate data and careful interpretation. If AI is being used to abstract clinical data, support quality reporting or measure performance, it needs to meet a high bar for accuracy, transparency and accountability.
To understand how healthcare organizations are approaching AI adoption in quality, my company recently surveyed hospital and health system quality leaders about their current use of AI in quality operations. The results point to a gap between AI enthusiasm and organizational readiness. Nearly nine in 10 respondents (88%) are “interested in using AI in healthcare quality operations,” yet only 23% consider their organizations “very ready to adopt it.” This finding is consistent with a 2025 survey from Nordic and Modern Healthcare, which reported a large disconnect between healthcare executives’ confidence in their organization’s AI readiness and how prepared those organizations actually were.
It’s clear that while interest in AI is widespread, many organizations still need to put the policies, processes and infrastructure in place to create the foundation AI needs for success.
Our survey found that a quarter of respondents have already implemented AI in quality operations, and another 30% are piloting, testing or evaluating AI solutions.
Quality leaders are most interested in using AI to help solve practical problems. The top motivation for using AI in quality was enhancing efficiency, cited by 57% of respondents. Deloitte’s latest global survey of healthcare executives points to the same priority: Operational efficiency ranked as AI’s most valuable use case in 2026. As for what specific quality functions respondents were most interested in supporting with AI, our survey found that the areas of greatest interest were quality or performance improvement, data analysis, data validation and quality assurance and clinical data abstraction.
Many of these uses require significant clinical judgment. Consider clinical data abstraction, for example. AI can efficiently identify inconsistencies across large volumes of data, saving time and reducing manual effort. But flagging inconsistencies is only part of the job. Clinical data is highly nuanced and requires interpretation before it can be abstracted, making human expertise an important part of the process. Exploring AI for these types of tasks requires careful consideration of where AI can add value, where human judgment remains essential and how the two can work together effectively.
That balance matters most as organizations consider expanding AI beyond pilots into broader quality operations. Respondents identified compliance concerns, trust in AI accuracy, costs and fragmented systems as their biggest barriers to AI adoption in quality.
The survey also points to a gap in AI governance. Only a third of respondents said their organizations have formal, documented AI governance policies. A 2025 report from the Healthcare Financial Management Association and Eliciting Insights found a similar gap, noting that while most health systems used AI internally, just 18% had “a mature governance structure and fully formed AI strategy.” As adoption moves beyond experimentation, governance will become increasingly important to ensure organizations have a consistent approach to evaluating AI opportunities, managing risk, defining appropriate uses and monitoring performance.
These challenges highlight the need for a thoughtful approach to AI adoption supported by the right infrastructure, processes and oversight.
Addressing the barriers to AI adoption starts with building the governance, people, systems and processes needed to use AI responsibly. Hospital and health system leaders can strengthen their readiness by focusing on a few key priorities.
Organizations need clear policies for evaluating AI tools, approving use cases, managing data, monitoring performance and assigning accountability—but most organizations we surveyed don’t yet have these formal, documented policies in place. Quality, clinical, compliance and technology leaders should all have a role in developing these policies. Policies should also establish how organizations will protect data, manage vendor relationships and meet compliance requirements.
Human expertise is essential when AI is applied to work that requires clinical or quality judgment. AI can surface relevant information and suggest potential answers, but human experts should review that information and make the final determination. For example, at my company, we’ve built this directly into abstractors’ workflow: Every suggestion includes its source evidence, and a clinical expert must review and approve it before it counts.
Data integration and interoperability are major barriers to scaling AI solutions. Breaking down data silos and centralizing clinical data management through one enterprise platform gives AI access to the volume and consistency of data it needs to identify patterns and produce more useful outputs.
Our survey found that 83% of respondents use or are considering third-party AI solutions, whether on their own or in combination with internally developed solutions. When evaluating AI vendors, organizations should look for solutions that make outputs easy to understand and review, such as by showing supporting evidence and allowing users to accept, modify or reject an AI-generated result. Features like this bring greater visibility into AI decision-making and help build trust in its accuracy.
AI has massive potential to change how healthcare organizations manage quality. But because this work requires clinical context and judgment, organizations will need to be thoughtful about how they introduce and expand its use.
The next phase of AI adoption will depend less on enthusiasm alone and more on organizational readiness. Healthcare organizations that establish clear governance, prioritize interoperability and maintain human oversight will be better positioned to move AI from isolated experiments into meaningful, scalable applications.
Forbes Technology Council is an invitation-only community for world-class CIOs, CTOs and technology executives. Do I qualify?
