Athenahealth Wants to Turn the EHR Into a Self-Learning System

“Rapid advancements in artificial intelligence are allowing us to revisit the role technology plays in healthcare… we are bringing customers along as we develop, test, and deploy solutions that will future-proof their practices,” said Bob Segert, chairman and CEO of athenahealth. Over the past year, athenahealth has moved from adding AI features toward re-architecting its entire EHR and revenue-cycle platform around intelligence. The company now describes its flagship system, athenaOne, as AI-native. From ambient transcription pilots to full revenue-cycle automation and payer-portal agents, the fight for EHR dominance is no longer just about features, but about who controls the data, the model and the feedback loop. The shift from add-on tools to embedded intelligence Earlier this year, athenahealth expanded its Ambient Notes program, partnering with transcription-AI vendors such as Abridge so that its network of over 160,000 clinicians could select ambient-scribe options within athenaOne. But the bigger step came with the announcement of in-house tools like athenaAmbient and the clinical copilot Sage, both designed to listen to clinician-patient conversations and immediately surface draft notes, diagnoses and orders. These capabilities mark a transition from automating documentation to fundamentally reworking the clinician experience. For example, the company says its next-generation athenaClinicals interface is designed to predict diagnoses (“clinically-inferred diagnoses”) and summarise major changes via an adaptive intelligence layer built on large-language models and rich ambulatory datasets. The clinical story is only part of it. Last month, the company also unveiled AI-native practice and revenue cycle (RCM) features built on a cloud-native, single-instance architecture. These include automated insurance selection, patient-liability estimation, wait-list scheduling, payer-portal agents and denial-advice automation. The results claimed are impressive: a 98.4% clean-claim rate, a median initial denial rate of 5.7%, and administrative work reductions of 50–70% for practices deploying the new RCM workflows. With a single-instance SaaS setup, every update goes to all customers, enabling rapid iteration and shared benefit. According to Bob Segert, “We can hit every single one of our 160,000-plus providers all at once when we release. That’s just a fundamental technological platform that’s really unique and differentiated.” Competing through data, not features While major EHR competitors such as Epic Systems and Oracle Health are also accelerating AI deployments, athenahealth’s pitch emphasises architecture and data network rather than discrete modules. Epic, for example, is expanding copilot and ambient-scribe tools across large health-system customers; Oracle’s next-gen EHR is built for voice, AI and cloud. What sets athenahealth apart is the shared code-base and network-wide intelligence. Because all customers are on the same instance of athenaOne, a model trained on a denial resolution in one practice can feed improvements to every other practice immediately. This network effect is rare in healthcare software, which is often fragmented and deployed locally. The company highlights that its AI-native platform is delivered through one cloud-instance to all customers. The company is rolling out many of its AI features at no additional cost to customers. By lowering the adoption barrier, athenahealth hopes to accelerate deployment and thereby accelerate data collection and learning. Of course, there are hurdles. Embedding ambient audio-capture in a clinical environment raises consent, privacy and medico-legal questions. Some practices will resist automation until the model’s accuracy and utility are proven. Plus, claiming 50-70% reduction in administrative workload and sub-6% denial rates is compelling, but real-world audits and long-term outcomes will matter. Regulatory scrutiny is likely to grow. As AI starts influencing diagnoses, orders and billing (domains embedded in care and payment workflows) the oversight that payers, regulators and professionals demand will intensify. Athenehealth’s leadership emphasises that data stays within its cloud, is anonymised in aggregation, and external LLMs are interchangeable rather than fixed, but trust must still be earned. For ambulatory practices, the opportunity is significant. Independent physicians and smaller practices face pressure from rising costs, declining reimbursements and clinician burnout. Athenehealth’s platform offers a path to reclaim time, reduce manual work and optimise cash-flow. “How do we enable these independent practices to thrive in a very difficult environment?” Segert asked .

Athenahealth Wants to Turn the EHR Into a Self-Learning System

The shift from add-on tools to embedded intelligence

Keep reading. It's free.

Related stories