Abridge Secures Eli Lilly Investment, Expands Beyond AI Scribe

Abridge landed a strategic investment from Eli Lilly on June 11, 2026, the same day the healthcare AI company unveiled a major expansion of its platform beyond clinical documentation into payer, claims, prior authorization, and clinical trial workflows. Financial terms of the investment were not disclosed.

The company now calls the new product an AI-native clinician intelligence platform, framing it as a step beyond the ambient medical transcription tool Abridge was founded to build eight years ago. At a midtown Manhattan keynote, CEO and co-founder Dr. Shiv Rao argued that the same clinical conversation that once generated a note can now drive billing, prior authorization, clinical trial recruitment, and evidence-based care.

Abridge Rebuilds Itself as a Clinical Intelligence Platform

The Times Center event on Thursday was, in effect, Abridge’s public reframe. Until this week the company sold itself primarily as a documentation tool, even as it spent two years quietly building out billing, coding, and decision-support features. Rao used the keynote to argue that the visit itself, captured in conversation, is now the right place to fix everything downstream.

We started Abridge to save time, save money, and save lives. This next chapter brings trusted intelligence into the most important moment in medicine: a clinician caring for a patient.

Dr. Shiv Rao, CEO and co-founder of Abridge, in a statement released with the keynote.

The new platform pulls pre-visit preparation, in-visit support, post-visit documentation, billing, coding, and clinical decision support into one workflow, the full scope of which Abridge laid out in Abridge’s June 11 platform announcement. It drafts a pre-visit summary tailored to the specialty and care setting, then surfaces discussion topics and answers clinician questions in natural language during the encounter. After the visit, it generates notes, flowsheets, patient summaries, billing codes, and orders for clinician review. Abridge says it is deeply integrated across Epic, Oracle Health, and athenahealth, the leading electronic health record (EHR) systems in the U.S. market, so documentation flows into the patient record without a separate data entry step.

• 300+ health systems live on Abridge
• 100M+ clinical conversations a year
• 250M+ patients served across those systems
• 28+ languages validated for speech recognition and note generation
• $830 million total raised since the company’s 2018 founding

Reid Health, a rural Indiana system, said Abridge has helped bring its nursing vacancy rate down from 18% to 8.6% with zero contract staff, and cut incidental overtime by 70% on the teams where it has rolled the platform out. Reid Health chief nursing officer Misti Foust-Cofield put it bluntly in a press statement: the tool gives nurses their presence back at the bedside. The numbers come from a single site and are self-reported, but they are the kind of metrics health system CFOs are now collecting.

Abridge has raised $830 million since 2018, including Abridge’s $300 million Series E funding round led by Andreessen Horowitz and joined by Khosla Ventures, making it one of the most heavily funded clinical AI developers in the country. The platform is the first AI-native clinician intelligence platform the company has shipped under that name, per Abridge’s announcement. The reframe is also the moment Abridge stops competing narrowly with other AI scribes and starts competing with the wider stack of revenue cycle, prior authorization, coding, and clinical research vendors that have lived in those lanes for years. The keynote gave that shift a deliberate line: “We’re building real-time bridges between patients, providers, and payers,” Rao said. Northwestern Medicine, the latest enterprise customer, has begun rolling Abridge out across every hospital and care setting in its network.

The strategic question now is whether the platform can monetize the new lanes the same way it has monetized documentation. Rao’s framing, that the clinical conversation is now the connective layer, only works if health systems and payers pay for that layer as a standalone contract, not as a feature inside a transcription subscription. Northwestern’s enterprise deal, the new American Heart Association, American Diabetes Association, and American Academy of Family Physicians content partnerships, and the Reid Health nursing metrics are evidence the platform is being used in production, not just in pilots. The market for that production, in revenue cycle, prior auth, and research, is where the company’s next growth bet is being placed.

Why Eli Lilly Is Betting on the Exam Room

Eli Lilly’s investment is small in disclosed dollar terms and large in strategic signal. The pharma company is betting that Abridge’s clinical conversation data can shorten the path from a new molecule to the right patient. The use case Lilly is funding first is trial identification: surfacing patients who may be eligible for the drugmaker’s clinical trials directly at the point of care.

Lilly executives did not disclose a dollar figure, and they did not need to, because the bet is about pipeline, not about returns in the next quarter. The life sciences module Abridge demonstrated at the Times Center pulls trial eligibility criteria from the patient summary and the running conversation, then routes flagged patients into screening pathways during the same encounter. Abridge’s own keynote page describes the clinical trial matching feature as built “all in partnership with Eli Lilly.” MedCity News first reported the investment and named that specific workflow as the focus. STAT News confirmed the deal the same day and added that Lilly was one of two trillion-dollar companies Abridge announced partnerships with.

Abridge is layering in additional research partnerships alongside the Lilly deal. The American Heart Association signed on to a parallel project to explore how ambient clinical intelligence could feed cardiovascular research and evidence generation. New content collaborations with the American Diabetes Association and the American Academy of Family Physicians will expand the company’s clinical decision support library.

When the latest evidence is available at the point of care, or at the point of need, it can save time, reduce variability, and support better outcomes for patients. Our responsibility is to help ensure that this science does not remain static but is integrated into the environments where decisions are being made.

Dr. Mariell Jessup, chief science and medical officer of the American Heart Association, in a statement released by Abridge.

Lilly is the first of the major pharma companies to back Abridge explicitly, according to MedCity’s reporting. The bottleneck in late-stage trials is patient identification, not molecule discovery, and the point-of-care conversation is one of the few places where eligibility signals sit in plain language. Other major pharma companies are now watching the early trial-enrollment numbers to see whether point-of-care trial screening works at scale. The same workflow is also why the AHA partnership is structured around research, not product integration.

Abridge Reaches for Real-Time Claims and Prior Auth

The bigger strategic swing at the Times Center was into the claims and prior authorization lanes. Abridge is now aligning real-time clinical documentation with real-time claims workflows, the kind of plumbing that decides whether a hospital gets paid, and whether a patient gets a procedure approved, in days rather than weeks. Aetna’s chief medical officer, Dr. Ben Kornitzer, and Johns Hopkins Health System CIO Deanna Hanisch appeared on stage with Rao to discuss how AI embedded in clinical workflows could give both sides the same picture at the moment of care.

Prior authorization is where the immediate customer pain is sharpest. At the J.P. Morgan Healthcare Conference in January, Abridge announced a partnership with real-time health information network Availity to bring AI-powered prior authorization to the point of care, with both companies saying the integration could speed prior auth from months to minutes, Abridge and Availity’s prior authorization collaboration. The company is also working with Highmark Health on a co-designed prior authorization tool built for the point of care. A year ago, Abridge introduced a contextual reasoning engine that produces billable notes supporting appropriate claims at the moment of capture, and that engine is now the substrate the new claims work is being built on.

For coding and clinical documentation improvement, Abridge is partnering with AHIMA, the national association for health information management. The collaboration is designed to support coding accuracy and auditability across both fee-for-service and value-based care reimbursement models, two reimbursement tracks that have historically pulled clinical documentation in opposite directions.

Rao described the goal as a bridge between providers and health plans, not a referee. “We’re building real-time bridges between patients, providers, and payers,” he said, framing the company’s role as a shared clinical record, not a side-taking claims reviewer. Cigna executives also appeared on stage at the Times Center to discuss the new workflow, suggesting at least two of the top U.S. insurers are watching the project closely.

The real-time claims vision depends on data-sharing and standards the healthcare system does not have today. Payer and provider systems still speak different languages on the back end, and the worst friction in claims adjudication happens after a patient has left the building. Abridge is positioning the documentation, generated from the conversation itself, as the common layer both sides can read. Whether that layer is enough to compress adjudication from weeks to minutes is the open question, and one the next twelve months of live deployments are designed to answer.

A Foundation Model Built Only for Clinical Conversations

The other half of the Times Center news was the partnership with Nvidia. The two companies are co-developing what Abridge and Nvidia call the first foundation model purpose-built for clinical conversations, a deliberate break from the dominant pattern in healthcare AI, where developers take a general-purpose model and fine-tune it on medical data. The model is being trained from the ground up on clinical conversations, with clinical knowledge applied at every training stage, Nvidia’s vice president of healthcare Kimberly Powell said on stage. The aim is a model that reasons like a clinician, not one that mimics a clinician. The model, when deployed, will sit underneath Abridge’s ambient documentation layer, the company said.

Powell described the gap in clinical AI in plain terms. Generic models, she said, do not understand clinical language, lack clinical reasoning, and lack the domain expertise to handle “all of the long-running tasks and interconnected work that has to happen for workflows to be completely transformed.” Nvidia’s argument is that a model trained only on clinical conversation is closer to a clinician’s actual workflow than a generalist retreaded for medicine.

Abridge and Nvidia are positioning the new model as a different category from the typical clinical AI fine-tune.

The technical claims will be tested against real workloads, including the high-volume ambient documentation Abridge already runs across its enterprise customers. That testing is the most likely near-term source of head-to-head benchmarks between the foundation model and competitors’ general-purpose offerings. For now, the foundation model is a roadmap commitment rather than a deployed product.

For Abridge, owning a purpose-built model also locks in a moat that other clinical AI developers will find hard to match without a comparable training dataset. The model, like the platform around it, will be evaluated on what it changes in real clinical and financial workflows, not on benchmark scores.

The Incumbents Abridge Is Now Sharing a Roof With

Abridge’s reframe puts it in the same rooms as a much larger set of incumbents. Revenue cycle management vendors, prior authorization platforms, coding services, and clinical trial recruitment companies have all owned lanes Abridge is now moving through. The strategy is to undercut them by owning the source of truth, the clinical conversation, rather than by building a better claim scrubber or a faster coding tool.

Each of those lanes has its own established vendors and its own procurement cycles. Revenue cycle management (RCM) is dominated by players that health systems already pay for claims automation, and prior authorization platforms have spent a decade integrating with both providers and payers. Coding and clinical documentation improvement, the lane Abridge is now moving into, is organized around professional associations, and Abridge has now partnered directly with AHIMA, one of the main bodies in the field. Clinical trial recruitment is a fragmented market that is still mostly human-staffed, and is the lane where Lilly’s investment lands. The new smart room integrations, with Artisight and hellocare.ai, push Abridge into inpatient virtual care, a category with its own active competitor set.

• Claims management: real-time claims alignment with payers, which Abridge says will compress reimbursement cycles
• Prior authorization: partnerships with Highmark Health and Availity, with Availity claiming prior auth could be reduced from months to minutes
• Coding and clinical documentation improvement: AHIMA partnership for accuracy and auditability across fee-for-service and value-based care
• Clinical trial screening: life sciences module that surfaces trial eligibility during patient encounters, the focus of the Lilly investment
• Smart room monitoring: integrations with Artisight and hellocare.ai for inpatient ambient documentation and virtual care

The strategic risk is that Abridge now has to perform in five lanes at once, each with its own compliance regime, sales motion, and integration burden. The opportunity is the opposite: a single ambient platform that becomes the substrate for all five workflows is hard to displace once it is wired into the EHR. Abridge says its platform is already deeply integrated with Epic, Oracle Health, and athenahealth, the leading EHR vendors in the U.S. market. The Reid Health nursing metrics, the Northwestern Medicine enterprise deal, and the 300-plus health system count are all evidence the integration is working in production, not just in pilots.

The next test is whether the platform can monetize the new lanes at the same scale it has monetized documentation. Rao’s framing, that the clinical conversation is now the connective layer, only works if health systems and payers are willing to pay for that layer as a standalone contract line, not as a feature inside a transcription subscription. Northwestern Medicine’s enterprise-wide deployment, announced the same day, is the first data point on that question, and the AHA, ADA, and AAFP content partnerships, also new, are part of the same bet.

Abridge has the funding, the customers, and the platform name to make the multi-lane play. What it does not yet have is proof that any single one of those lanes, claims, prior auth, coding, trial recruitment, or inpatient smart rooms, can scale outside the company’s existing footprint.

The Open Questions for Abridge’s New Stack

Three things remain unproven about the platform Abridge unveiled at the Times Center. The first is unit economics: the company has not disclosed the pricing for the new claims, prior auth, or research modules, and the strategic investments from Eli Lilly and Nvidia, both of which are strategic rather than purely financial, do not reset the price customers will pay. The second is real-time adjudication: Abridge and Availity said in January the integration could compress prior auth timelines by orders of magnitude, and that claim has yet to be independently verified at scale.

The third open question is competitive. Abridge is moving into lanes already owned by established vendors, and the company’s own announcement included Aetna, Cigna, Johns Hopkins, and Reid Health as named partners across those lanes. Each of those partners is also a customer of at least one of the incumbents Abridge is positioning against, and the early deployments in the next 12 months are the test case for whether the ambient platform can replace those point solutions or only layer on top of them. The first signals will come from how much of Abridge’s new revenue, on top of the $830 million it has already raised since 2018, comes from the documentation lane versus the new ones.

Frequently Asked Questions

What is Abridge’s new platform?

Abridge’s expanded product, which the company calls an AI-native clinician intelligence platform, was unveiled at a Times Center keynote in New York. It connects pre-visit preparation, in-visit support, post-visit documentation, billing, coding, and clinical decision support inside a single workflow, and reaches into payer, prior authorization, and clinical research lanes alongside documentation.

Why is Eli Lilly investing in Abridge?

Eli Lilly made a strategic investment in Abridge on June 11, 2026, to support evidence-based care and research. The deal centers on Abridge’s life sciences module, which can surface clinical trial eligibility during a patient encounter and route flagged patients into screening pathways. Financial terms were not disclosed.

How does the Nvidia partnership differ from existing medical AI?

Abridge and Nvidia are co-developing what they call the first foundation model purpose-built for clinical conversations, applying clinical knowledge at every training stage rather than fine-tuning a general-purpose model on medical data. The goal, per Nvidia’s Kimberly Powell, is a model that reasons like a clinician, not one that mimics one.

What will real-time claims workflows do for healthcare?

Abridge is aligning real-time clinical documentation with real-time claims processing, a workflow the company says will let health systems and payers reduce administrative rework and accelerate reimbursement. The full vision, real-time adjudication, is the open test for the platform in 2026 and 2027.

How is Abridge being used by major health systems?

Abridge is live at more than 300 health systems, including community health centers, specialty hospitals, and large academic systems, supporting more than 100 million clinical conversations a year across organizations that collectively serve more than 250 million patients. Northwestern Medicine is the latest enterprise customer, with a deployment across all of its hospitals and care settings.