Inpatient AI, Payor Win-Wins, and Token Costs (with Dr. Shiv Rao, co-founder and CEO of Abridge)
Tuesday, October 20, 2026
Transcript intelligence
What OpenRounds heard
Source-grounded signals extracted from the episode transcript. Timestamps and supporting excerpts appear when available.
The speaker reports that Abridge leveraged playbooks and engines developed from outpatient and emergency room deployments to accelerate inpatient adoption, and is currently experimenting with an agent that provides decision support on inpatient versus observation status criteria.
The speaker reports that AI native companies can differentiate by distilling, mid-training, and post-training proprietary models to own and control compute, allowing them to smartly match model size to problem size and ride frontier models, thereby providing health system customers with psychological security and certainty around delivered value and predictable costs rather than open-ended tokenomics.
Hospitalogy’s Blake Madden sits down with Dr. Shiv Rao, CEO and co-founder of Abridge, two hours before Rao was going on call at the hospital. They talk about how Abridge has taken its playbook from ambulatory to the ER and now inpatient, and what’s different about engineering for that setting, why AI should rethink the claims process rather than be sprinkled on it, and how Abridge is working to help health systems avoid unpredictable token costs. Also in this episode: What happens to a claim after a doctor types “shortness of breath” into a note? Rao says the whole loop adds up to roughly $300 billion of waste. What separates real products from “party trick” prototypes? A University of Vermont case shows the value-add of context engineering, guardrails, and knowing what good looks like. One payor told Abridge it approves well over 90% of the prior auths it initially denies and would welcome automation. Rao says scale earned Abridge a seat with payors, and he wants to avoid a future of “agents fighting agents.” Assort Health’s AI agents complete the work that turns patient demand into care, including scheduling, intake, referrals, forms, medication refills, and payments. Customers have increased appointment volume by 5%, expanded staff capacity by 115%, and captured up to $4.3 million in new annual revenue per 100 providers. To learn more, check them out today at https://www.assorthealth.com/hospitalogy%E2%81%A0%E2%81%A0 For expert analysis on healthcare M&A, strategy, finance, and markets, join 80K+ readers of Blake Madden’s Hospitalogy newsletter. Subscribe here: https://hospitalogy.com/%E2%81%A0%E2%81%A0%E2%81%A0%E2%81%A0%E2%81%A0%E2%81%A0%E2%81%A0%E2%81%A0%E2%81%A0%E2%81%A0 Like our podcast? Listen to previous episodes of Claims Denied here: https://hospitalogy.com/podcast/%E2%81%A0%E2%81%A0%E2%81%A0%E2%81%A0%E2%81%A0%E2%81%A0%E2%81%A0%E2%81%A0%E2%81%A0%E2%81%A0 Be sure to leave us a review, and share our podcast with colleagues. Follow Blake Madden on LinkedIn: https://www.linkedin.com/in/blakecmadden/%E2%81%A0%E2%81%A0%E2%81%A0%E2%81%A0%E2%81%A0%E2%81%A0%E2%81%A0%E2%81%A0%E2%81%A0%E2%81%A0%E2%81%A0%E2%81%A0
Show transcript excerpt
Hospitalogy’s Blake Madden sits down with Dr. Shiv Rao, CEO and co-founder of Abridge, two hours before Rao was going on call at the hospital. They talk about how Abridge has taken its playbook from ambulatory to the ER and now inpatient, and what’s different about engineering for that setting, why AI should rethink the claims process rather than be sprinkled on it, and how Abridge is working to help health systems avoid unpredictable token costs. Also in this episode: What happens to a claim after a doctor types “shortness of breath” into a note? Rao says the whole loop adds up to roughly $300 billion of waste. What separates real products from “party trick” prototypes? A University of Vermont case shows the value-add of context engineering, guardrails, and knowing what good looks like. One payor told Abridge it approves well over 90% of the prior auths it initially denies and would welcome automation. Rao says scale earned Abridge a seat with payors, and he wants to avoid a future of “agents fighting agents.” --- Assort Health’s AI agents complete the work that turns patient demand into care, including scheduling, intake, referrals, forms, medication refills, and payments. Customers have increased appointment volume by 5%, expanded staff capacity by 115%, and captured up to $4.3 million in new annual revenue per 100 providers. To learn more, check them out today at https://www.assorthealth.com/hospitalogy --- For expert analysis on healthcare M&A, strategy, finance, and markets, join 80K+ readers of Blake Madden’s Hospitalogy newsletter. Subscribe here: https://hospitalogy.com/ Like our podcast? Listen to previous episodes of Claims Denied here: https://hospitalogy.com/podcast/ Be sure to leave us a review, and share our podcast with colleagues. --- Follow Blake Madden on LinkedIn: https://www.linkedin.com/in/blakecmadden/
Daily healthcare AI brief
Get the free healthcare AI briefing
One concise briefing built from reporting, research, policy, blogs, videos, and full podcast transcripts.
Free. Get a sign-in link by email to finish subscribing. An account is created if needed. Unsubscribe anytime. Already subscribed? Sign in to save and personalize.