Navigating AI’s growing role in healthcare plans

By Charity Gourley

Whether you like AI, don’t like it, find it useful or not, one thing is clear: AI is here to stay. There is a rapid proliferation of AI tools, bots, and systems in all areas of life, but especially in how many people receive and engage with healthcare. When Trusts are choosing a vendor or plan provider, it can help Trustees to be aware of the many ways AI can show up, and how it might impact their plans and participants.

Here are a few common functions of AI in healthcare plans:

Benefits Navigation & Personalized Guidance  

Generative AI – or artificial intelligence that creates new original content rather than just analyzing or categorizing data – offers participants a way to interact with their health benefits in new and increasingly sophisticated ways. AI can be purposefully built and designed to improve healthcare navigation by generating personalized information based on a participant’s plan, network, and coverage. This allows plan providers and vendors to better guide participants to appropriate resources and care solutions, which in turn can help improve utilization and reduce plan costs.

Workflow Automation

A recent study conducted by Transcarent found that 91% of HR leaders reported getting frequent and repetitive questions about benefits and that 85% spend significant time on documentation and compliance. While the study focuses on HR departments, there is a similar trend in plan administration with a large amount of administrative time going to participant questions, vendor coordination and managing compliance. Clinical-grade and purpose-built AI solutions can alleviate some administrative burdens and ease compliance challenges by:

• Operating as a first tier to answer participant questions with plan-specific answers, allowing human support teams to focus on more complex concerns
• Automating clinical notes and claim reviews to speed up processing times and flag important or missing information
• Funneling inquiries or escalations to the right departments or team members
• Supporting HIPAA compliance, SOC 2 certification, and safeguards for privacy, bias, and regulation

Virtual Health Assistants

In an April 2025 study it was found that 35% of Americans could not access quality health care if they need it. That equates to around 91 million people struggling to access care. AI-powered virtual assistants and virtual care platforms can improve access to care, especially for people in rural areas. Many plans and wellness vendors offer some form of virtual assistance to help participants navigate appointment scheduling, get basic health guidance, connect people with in-network healthcare professionals, and limit barriers to care such as travel time and wait lists. More advanced AI platforms are designed to offer more nuanced health guidance and to incorporate data from fitness trackers, sleep monitors, nutrition apps, and medical records to provide a comprehensive view and offer preventive solutions. Virtual assistants and platforms can offer real-time and personal support, making mental and physical health care more accessible, immediate, and integrated.

Questions To Ask About AI Platforms

When evaluating a vendor or provider plan’s AI tools and resources, a few key questions to ask include:

• What functions are the AI performing – reviewing claims and prior authorizations, or serving as customer service, etc.?
• How is the efficacy of AI tools measured and demonstrated?
• Are platforms and systems monitored, secure and compliant?
• What tools make healthcare information more accessible to participants?
• How are AI systems designed to reduce administrative burdens?
• What is the review or appeal process for AI-based decisions?

While AI tools — from generative chatbots to virtual health assistants — are rapidly becoming part of the healthcare benefits landscape, it’s important for unions and plan sponsors to understand both their potential and limitations. AI can support benefits navigation, reduce administrative burden, and help members make more informed decisions, but AI does not replace the need for robust safeguards around privacy, accuracy, bias mitigation, and human oversight. Asking vendors clear, targeted questions about how their AI systems operate, how they are monitored, and their impact on participants will help ensure that these technologies strengthen plan value and protect member interests as adoption continues to grow.