The US healthcare industry spends at least $82.7B a year on back-office administrative work. In 2022, spending on administrative tasks increased a staggering 50% compared to the previous year. To put it plainly: this is a massive, growing problem for an ecosystem already plagued by significant challenges.
Despite this, a tremendous amount of healthcare data is exchanged via manual phone calls between payors, providers, patients, and pharma companies. Some of these calls are to manage prior authorization of future medical procedures, while others are to manage dispute resolutions between various parties—like who’s going to pay for medical services already delivered. Most of these calls are inefficient, since most parties spend their time waiting on hold instead of serving patients.
Estimates suggest that every year, billions of these back-office calls take place, and last 20–30 minutes. Such operational inefficiencies may seem run of the mill, but they can cause delays in patient access to treatment and lead to even more expenditures in the future—not to mention increased rates of agent burnout in call centers, which have their own costly downstream impacts to the ecosystem. And according to the most recent CAQH Index, time spent on such administrative work increased 14% on average year over year.
Infinitus is tackling this problem by building the first AI platform specifically designed to automate healthcare phone calls and data gathering. In doing so, the company is meaningfully accelerating patient access to care while mitigating costly employee turnover and improving data capture quality.
The Infinitus AI agent handles phone calls end-to-end on behalf of healthcare providers, fully automating tasks like benefit verification that can last up to an hour for human callers, and updating patient records in real time. Their new proprietary AI copilot, FastTrack, helps front- and back-office staff in every part of healthcare navigate complex, legacy interactive voice response (IVR) systems, and waits on hold on their behalf before gracefully dropping callers into a conversation once a live payor agent is available. Both solutions draw from a vast knowledge graph of payor intelligence gathered from over 4M completed calls.
Infinitus is also pioneering the safe and accurate use of AI in patient care. Unsurprisingly, healthcare-related phone calls can be lengthy and complex, and the bar for accuracy is exceedingly high. To meet this need, Infinitus has built strict AI guardrails through a coordination layer that sits on top of LLMs and restricts conversations to an approved set of topics while still allowing for bespoke conversational turns and long context windows—a capability even the most modern LLMs have yet to achieve. This makes possible the adoption of AI without fear of hallucinations or bias, a landmark achievement for healthcare.
This is one of the most obvious—yet interesting and computationally complex—use cases for AI in a world where people often ask about the technology’s real-world business applications. What is AI really good for? Parsing unstructured data in lots of disparate documents (in this case, payor policies, formularies, and patient coverage documents) to make it easier for users to answer a simple question, like: will my insurance company reimburse my medical procedure?
Infinitus is poised to transform how the healthcare ecosystem exchanges information, helping to accurately and efficiently process the massive amount of data that is integral to getting patients high-quality care.
We first met cofounders Ankit Jain and Shyam Rajagopalan before the company was incorporated, and it’s been inspiring to watch them tackle such a sweeping problem that impacts every single healthcare institution. We are thrilled to have led Infinitus’ Series C and to have joined the board.
