Somewhere in the United States right now, a billing specialist is on hold with an insurance company, waiting to ask a question whose answer already sits on a screen. Multiply that call by millions and you get one of the most expensive habits in American healthcare. McKinsey estimated that simplifying healthcare administration could save up to $265 billion a year, more than a quarter of the roughly $950 billion the country spent on administration in 2019.

Automation is already chipping away at that bill. Electronic transactions helped US healthcare avoid $258 billion in administrative costs in 2024, while more than half of health plans now use AI in their administrative workflows. A new problem is emerging as a result. AI agents now sit on both ends of the phone line, with providers' bots calling payers' bots to ask for information that lives in a database.

SuperDial, whose AI agents handle payer communication for revenue cycle teams, has now completed more than 10 million AI-powered calls to insurers. The milestone has led the company to an unusual conclusion for a business built on automating phone calls: the next breakthrough is making those calls unnecessary. On 17th November, it introduced Agent-to-Agent RCM, an approach that measures success by resolved requests, not completed calls.

A Quarter-Trillion-Dollar Admin Problem

McKinsey's estimate breaks the $265 billion opportunity into three layers. About $175 billion could come from changes inside individual organizations, $35 billion from better coordination between organizations and $105 billion from industry-wide changes such as shared standards, with some overlap between the last two. That means about two-thirds of the savings sits within the reach of single providers and payers acting on their own. (our analysis)

McKinsey estimated $265 billion in annual savings from simplifying US healthcare administration, against $950 billion of administrative spending.

Much of that spending flows through revenue cycle management, or RCM, the work of turning patient care into payment. It covers checking a patient's eligibility and benefits, securing prior authorization, following up on claims and keeping provider records current with each insurer. Every one of those steps requires information from a payer. Much of it still moves by phone, fax and web portal.

Automation Works as AI Arrives

The 2025 CAQH Index, which tracks administrative transactions across more than 600 provider organizations and health plans covering 63% of insured lives, shows the industry is getting faster. Electronic transactions helped avoid $258 billion in administrative costs in 2024, up 17%, while medical administrative spending fell 9%. Electronic prior authorization rose from 31% to 40% between the 2023 and 2025 Index reports, while dental claim payments moved from 21% to 33%.

Electronic transactions helped US healthcare avoid $258 billion of administrative costs in 2024.

AI is accelerating that shift. More than half of health plans and a quarter of provider organizations now use AI tools in administrative workflows, while about two in three physicians used AI in some form by early 2025. CAQH still sees about $21 billion in savings available from automating the transactions that remain manual or partly manual.

Paperwork Between Patients and Care

The burden falls hardest on clinics. The American Medical Association's latest survey of 1,000 physicians found that practices handle an average of 40 prior authorizations a week, consuming 13 hours of physician and staff time, or about 1.6 working days (our arithmetic). Two in five practices employ staff who work only on prior authorization.

Physicians handle an average of 40 prior authorizations a week, taking 13 hours of physician and staff time.

The cost reaches patients. 95% of physicians say prior authorization delays access to necessary care, while 79% report that patients abandon treatment because of authorization hurdles. Six in ten physicians also worry that AI on the payer side could raise denial rates, a sign that automation on one side of the system does not automatically help the other.

10 Million AI Calls

SuperDial's AI agents place calls on behalf of providers and billing companies, navigating phone trees, waiting on hold and talking to payer representatives to retrieve benefits, prior authorization and claim information. The company passed more than 1 million calls at the time of its $15 million Series A in June 2025, led by SignalFire with Slow Ventures, BoxGroup and Scrub Capital. It has now crossed 10 million, a tenfold rise in about 17 months (our analysis).

The platform serves provider groups, health systems, RCM companies and payers. It automates eligibility and benefits verification, prior authorization, claims follow-up, credentialing, enrollment and provider data management across phone, payer portals, EDI, APIs, fax and documents, then returns structured results to customers' systems. West Coast Dental has used it for more than 10,000 claim-status calls a month, while billing teams have reported up to 4x productivity gains.

SuperDial passed 10 million AI calls to insurers in November 2026, up from more than 1 million at its June 2025 Series A.

Bots on Both Ends of the Line

AI has arrived on both sides of the transaction. Providers deploy agents to chase claim and benefit information, while payers deploy automation to answer those requests. Each side is getting better at keeping the exchange going. Neither is getting better at ending it, which risks automating the very inefficiency the industry set out to remove.

SuperDial's answer is to stop treating the phone call as the unit of work. If an agent already has information, that context should travel with the request. If a digital channel can supply the answer, there should be no call at all. The company's goal is fewer exchanges, not faster ones.

Agent-to-Agent RCM

Under Agent-to-Agent RCM, SuperDial takes responsibility for resolving a customer's request across every available channel, so the customer does not have to manage each one. If an electronic transaction can answer the question, SuperDial uses it. If the information sits in a payer portal or can be pulled by faxback, its agents retrieve it there. When a phone call is unavoidable, the agent carries forward what is already known and asks only for what is missing.

Agent-to-Agent RCM checks electronic, portal and faxback channels before making a call.

The model changes how the value of AI is measured. Most automation vendors count the calls an agent completes, while SuperDial wants to count the work it removes. The platform maintains controls designed to support HIPAA compliance, the US law that protects patient health information. It has also completed a SOC 2 Type II examination, an independent audit of its security controls.

Four Minutes When AI Meets AI

The early data shows what happens when agents talk to agents. Among successful requests where SuperDial retrieved the information it needed, its agents averaged about four minutes per call attempt when the payer used voice automation, against seventeen minutes or more when they reached a human representative.

For successful requests, SuperDial's agents average about 4 minutes per call attempt with payer voice automation, against 17 minutes or more with a human representative.

That makes an AI-to-AI exchange more than four times faster, saving about 13 minutes, or 76%, of each successful attempt. SuperDial says the approach is already helping teams obtain payer information faster with less manual effort. It sees these automated exchanges as an early look at what direct agent-to-agent communication could deliver.

The Company Behind the Calls

SuperDial was founded by Stanford graduates Sam Schwager, now chief executive, with Harrison Caruthers. The pair previously ran a healthcare billing company and spent thousands of hours on repetitive payer calls. The company began as SuperBill and moved into voice automation in late 2023, then reached seven-figure revenue soon after launch.

It has raised more than $20 million in total, including the $15 million Series A that combined equity with $3 million of venture debt. Its customers include RCM companies and large provider organizations such as dental support organizations and management services organizations.

The Next Billion Requests

SuperDial's long-term goal is an AI-native network for payer-provider communication, where agents exchange trusted information directly without repeatedly requesting it across disconnected systems. For payers, that means fewer repetitive inquiries reaching their staff. For providers, it means fewer hours spent chasing information, while patients face fewer administrative delays between them and care.

Ten million calls have shown that AI can take on healthcare's phone burden. SuperDial now wants to power the next billion payer requests through Agent-to-Agent resolution, in a market where McKinsey sees $265 billion a year in savings and where the best call is increasingly the one nobody has to make.

Don’t forget to like and share the story!

Vested Interest Disclosure: HackerNoon has reviewed the report for quality, but the claims herein belong to the author. #DYOR.