More than 230 million people use ChatGPT to ask health and wellness questions every week. Meanwhile, four out of five physicians are considering using AI in their practices, and Claude for Healthcare, a large-scale language model launched in January, is poised to transform medical billing.
As new artificial intelligence (AI) capabilities in healthcare emerge every day, healthcare leaders are faced with the paradox of urgency to implement and persistent reluctance to scale up.
There are also signs that AI could improve access to medical information and guidance for vulnerable populations, with 600,000 health-related queries to ChatGPT coming from underserved rural areas each week.
However, rising costs of capital continue to reduce the availability of funds for strategic investments. As a result, AI use cases in healthcare often struggle to get past the pilot stage. In this day and age, making the right bet on AI in healthcare depends on a responsible AI strategy and a willingness to move quickly to seize opportunities at scale.
Beyond AI experiments
AI adoption in healthcare increased more than twice the rate of adoption in other industries from 2023 to 2025.
Optimizing the business value of AI in healthcare depends not only on deploying the right tools in the right areas, but also on scaling these tools quickly. They also need to think carefully about how to leverage AI quickly without putting undue stress on already challenged employees.
For example, agent AI holds great promise for reducing administrative burden and costs in the healthcare revenue cycle, but there are barriers to its adoption across the industry. One of the biggest obstacles is the fear of being replaced by AI. Two out of three leaders surveyed say this concern is contributing to their hesitancy to adopt agentic AI.
And while AI use cases are growing in healthcare, KLAS Research analysis shows that most organizations are using AI for lower-risk purposes, such as ambient AI to generate structured notes from doctor-patient conversations, image triage, predictive risk modeling, and responses to patient messages.
Move the needle on AI ROI
With limited resources available to invest in AI, how can healthcare leaders take the right actions to realize short- and long-term value? Here are three considerations in a transformative environment.
-
Don’t just taste the tools. chase the tools. Capital constraints require a shift from experimentation to implementation. Rather than testing individual tools, organizations should prioritize solutions that are designed to scale from the beginning. For healthcare leaders, this means looking for tools that are built for one workflow or hospital but can seamlessly migrate to other workflows or hospitals. This approach accelerates adoption, reduces development costs, and strengthens the foundation for data governance and employee adoption.
When deciding which AI initiatives to pursue, it’s also important to map their value to your healthcare organization’s strategy. This helps avoid isolated AI use cases that struggle to deliver their intended value. It also ensures that the investments an organization makes in AI are carefully aligned with the health system’s mission, vision, and business objectives, carefully balancing risk and reward.
-
Make sure the tool you choose is compatible with your EHR. This budget-protective approach paves the way for rapid innovation through seamless integration. It also makes it easier for team members to engage with AI tools by providing easier access and integration with other features. The ability to draw from EHR data also provides the foundation for actionable intelligence based on clean, controlled, and accessible data. With this foundation, healthcare teams can more easily turn problems into actionable innovations with great potential for ROI.
In 2026, key opportunities to connect AI innovation to EHR integration include:
- Improving clinical workflowincluding documentation, imaging, and diagnosis.
- patient involvementEnhance patient interactions with virtual assistants, AI-powered symptom checkers, educational resources, and more.
- revenue cycle managementAI can transform unstructured data into usable information to streamline pre-approvals, billing, coding, and denials.
- Control to monitor AI value. A recent EY Global Risk study found that organizations that monitor the performance of their AI initiatives in real time are 65% more likely to achieve cost savings. (Editor’s note: The author of this article is affiliated with EY). This highlights the importance of properly embedding AI within the broader digital ecosystem and regularly tracking AI efforts with the support of an AI governance committee. Committees should focus on value metrics such as operational efficiency, quality improvement, cost savings, and patient outcomes to ensure that AI delivers measurable benefits and is aligned with organizational goals.
As the pace of AI adoption in healthcare increases, the driving force is to develop an implementation roadmap that emphasizes rapid adoption and benefits while mitigating risks. It also increases team members’ confidence in innovation. This is essential to sustaining AI progress over the long term.
John Ward is a Technology Consulting Partner and Head of Health Technology at EY.
