Mapping Markets
April 16, 2024

Payer-Facing AI Phone Calls Market Map

Patrick Wingo's headshot
Patrick Wingo
Head of Research, Elion

This is part of Elions weekly market map series where we break down critical vendor categories and the key players in them. For more, become a member and sign up for our email here.

Our focus this week is at the center of gen AI x RCM. We’re seeing payers get tougher when it comes to prior authorization and claims denials, which is driving more and more back-office work for providers. Many HCOs, especially virtual-first and brick-and-mortar clinics, are looking to AI agents to even the playing field. Enter Payer-Facing AI Phone Calls with use cases like:

  • Prior authorization

  • Eligibility checks and verification of benefits

  • Claims status checks

  • Denied claims appeals

  • Credentialing and provider management

Conversations with payer IVR systems and call centers are generally more structured and forgiving of mistakes than patient-facing conversations. However, the large quantity of data collected through these calls and the importance of that data being correct means that providers need to consider the following:

  • Overall Technical Architecture, i.e. does the product use purely intent-based call flows with commodity off-the-shelf solutions, or does it intelligently incorporate LLMs for conversational flexibility?

  • Observability and Monitoring to help identify issues, improve calls over time, and ensure that data from the calls is correctly captured and labeled

  • Data Access Capabilities, including API and database access to look up records and incorporate them into the conversation as necessary

Products we’ve encountered in this space and seen more substantial demand for in our marketplace include Infinitus Eva, SuperDial, Amperos, and Opkit.

These spaces are new, and we expect to see many more use cases and companies emerge. We’re also chuckling at the implications of AI phone systems becoming prevalent: will outbound AI agents force payers to respond with their own inbound-focused AI? At what point will AI agents realize they’re talking to another AI, and will the conversation change as a result?

Related categories: