The use of artificial intelligence and large language models is progressing rapidly across the pharmaceutical industry, including within Market Access functions. Less is known, however, about the extent to which the payers are themselves adopting these technologies.
To better understand this evolution, Elevio Group conducted its first Payer AI/LLM Barometer in October 2026. The research combined a quantitative survey of 24 payers in the US and Europe with qualitative interviews with ten payers across the US, France, Germany, Japan, and China. The analysis covered six areas of payer activity: landscape and disease understanding; evidence search and synthesis; evidence assessment and analytics; economic and budget impact assessment; pricing, reimbursement and decision support; and reporting and workflow support.
The findings indicate that payer adoption of AI is already more advanced than might be assumed, although it remains highly heterogeneous across geographies and use cases. They also suggest that the areas in which AI is most frequently used today are not necessarily those where payers expect it to create the greatest value in the future.
AI is already becoming part of routine payer activities
The most established applications of AI are currently found in evidence search, synthesis, and information-intensive workflow activities. Across respondents, 61% reported routine use of AI for literature searches and evidence surveillance, while 50% reported routine use for the summarization of clinical or economic evidence. Routine use was also reported by 61% of respondents for administrative support, classification, and data extraction.
These use cases are relatively intuitive starting points for payer organizations. They involve large volumes of information, can consume significant professional time, and generally allow the output generated by AI to be reviewed before it influences a formal assessment or reimbursement decision. Qualitative interviews similarly identified evidence search as an early adoption opportunity because of its relatively limited risk compared with applications closer to formal decision-making.
The survey nevertheless shows that AI is already being used beyond administrative or literature-related activities. Among respondents, routine use was reported for activities including the identification of evidence gaps and uncertainty, comparative clinical effectiveness, analysis of claims or real-world evidence, and the preparation of assessment reports. The level of adoption differs considerably depending on the geography and the organization.
A significant adoption gap is emerging between US and European payers
One of the clearest findings of the barometer is the difference between US and European respondents. US payers reported considerably more routine use of AI across several activities related to disease understanding and evidence analysis.
For disease education and patient pathway understanding, for example, 60% of US respondents reported routine use, compared with 17% of European respondents. For the review of treatment guidelines and standards of care, the corresponding figures were 70% and 14%. A similar difference can be observed in evidence search, with 70% of US respondents routinely using AI for literature search and evidence surveillance compared with 43% of European respondents.
The qualitative interviews provide some context for this difference. Several US organizations appear able to draw on significant internal technology resources and, in some cases, dedicated AI capabilities. European payer and HTA organizations more frequently highlighted constraints associated with technical resources, confidential information, legal frameworks, and access to solutions that can be deployed within an appropriate secure environment. The difference observed in the survey therefore appears to reflect not only different attitudes toward AI, but also differences in the ability of organizations to deploy and govern these technologies.
This distinction will be important to monitor over time. European payers do not necessarily attribute less potential value to AI; in several areas, their expectations for future use are as high as, or higher than, those observed in the US.
Future value is expected to move closer to evidence assessment and reimbursement decisions
The largest opportunity identified by respondents extends beyond the activities in which AI is most widely used today. Payers see significant future potential in using AI to support more complex analytical activities, including the identification of evidence gaps and uncertainty, assessment of clinical evidence, review of health economic models, and preparation for reimbursement decisions.
This is particularly visible among European respondents. While only a minority currently use AI routinely to review manufacturer submissions or identify evidence gaps and uncertainty, 80% rated the future value of each of these applications as high, with an additional 20% considering evidence-gap analysis potentially transformative. For comparative clinical effectiveness, 60% rated future value as high and 40% as transformative.
Expectations are also high for economic assessment. Among European respondents, 60% considered AI potentially transformative for the review or quality control of health economic models, and 40% did so for budget-impact analyses. The qualitative interviews suggest that resource constraints may be an important driver of this interest: public payer organizations are facing increasing analytical workloads while access to specialized health economic expertise can remain limited.
In the US, the expected applications extend further into pricing, coverage, and reimbursement decision support. Seventy percent of US respondents rated the potential value of AI in preparing pricing or reimbursement negotiations as high, while 64% rated its potential value in comparing coverage, reimbursement, and prior-authorization scenarios as high.
These findings suggest a progressive evolution in payer AI use. The first generation of applications has largely focused on accelerating access to information and reducing the time required for synthesis and administrative work. Payers increasingly envisage AI supporting activities that require interpretation of evidence, identification of uncertainty, testing of alternatives, and preparation of recommendations.
What does this mean for pharmaceutical Market Access teams?
As payer adoption progresses, pharmaceutical companies will increasingly need to consider AI not only as a tool for improving their own Market Access activities, but also as a factor influencing how external stakeholders review and interrogate evidence.
If payer organizations progressively use AI to identify evidence gaps, review economic models, compare treatment alternatives, or prepare reimbursement scenarios, this may change the speed and depth with which evidence packages can be assessed. It may also place greater emphasis on the consistency, transparency, and accessibility of the underlying evidence submitted by manufacturers.
Elevio Group's 2026 Payer AI/LLM Barometer provides the detailed results across the six domains assessed, including US and European comparisons and qualitative perspectives from the US, France, Germany, China, and Japan.
For Market Access leaders interested in the full findings, contact us to access the complete barometer read-out and discuss the implications for evidence, access strategy, and engagement with payers.
About Elevio Group
Elevio Group is a global Market Access strategy consultancy supporting pharmaceutical and biotech companies throughout drug development and commercialization. We combine deep Market Access expertise, strategic consulting, technology, and AI-enabled innovation to help clients strengthen evidence strategies, optimize market access, and maximize the value of their assets.
