Abridge Keynote | 6.11.26 | New York City

From the Times Center in New York City on June 6, 2026: Abridge CEO and Co-Founder Dr. Shiv Rao announces the patient-centered clinician intelligence platform—a single, horizontal intelligence layer, connecting providers, payers, and life sciences, organized around every patient, for every moment of care. 00:06:15 — Northwestern Medicine Partnership Announcement Shiv announces that Northwestern Medicine is now implementing Abridge enterprise-wide. 00:11:44 — Clinician Intelligence Platform Announcement Announcing the new Abridge Patient-Centered Clinician Intelligence Platform. Dr. Shiv Rao walks through the full scope: compliance, billing, clinical intelligence, and decision support, and unveils a wave of new collaborations with the American Heart Association, AHIMA, AAFP, ADA, and leading medical journals. 00:12:30 — Eli Lilly Strategic Investment Eli Lilly & Co. is making a strategic investment in Abridge to support evidence-based care and research access at the moment of care. 00:16:45 — Inpatient Demo AI capture of the clinician-patient conversation — no screens, no typing. In-house, medically-tuned speech recognition models built to work even in noisy hospital environments. 00:17:57 — Pre-Visit Note Generation Before the clinician ever walks into the room, Abridge generates a pre-visit note pulling from the EHR, prior clinician notes, and (in the outpatient case) the entire inpatient stay, delivering continuity across care settings rather than starting fresh each visit. Surfaces a clean summary along with auto-generated discussion topics. 00:20:31 — Background Agent Reconciliation A team of agents works silently behind the scenes: reading prior notes and labs, extracting diagnoses, reconciling against payer medical necessity criteria, and checking diagnosis ontologies. Surfaces only as discussion topics in the clinician's workflow, explicitly, "no alerts, no pop-ups.” 00:22:30 — Documentation Integrity Post-visit, a low-latency model detects documentation gaps and auto-generates guideline-based queries to close them. It's designed to reduce denials at the source rather than fight them after the fact. Three-tier model: org-level compliance floor → specialty-level style → individual clinician's own documentation voice, trained on their past notes. 00:27:35 — Outpatient Demo Ten days post-discharge, the patient is with her PCP visit. The demo shows how the discharge-to-outpatient gap is closed because the pre-visit note already carries the full inpatient context forward. 00:30:01 — Risk Adjustment and Coding Dr. Shiv Rao reveals the in-house model converting diagnoses to ICD-10 codes in under 600ms, with an 85% F1 score vs. 69% for a frontier general-purpose model. The neuropathy finding from the inpatient stay carries through and gets coded correctly here. 00:31:11 — Clinical Trial Matching Abridge surfaces a relevant clinical trial based on the patient's history, ejection fraction, and guideline-directed therapy status, reasoning across the clinical trial management system, EHR, and payer data simultaneously. 00:36:41 — Panel: AI-Enabled Value-Based Care Deanna Hanisch, CIO of Johns Hopkins Health System, and Dr. Ben Kornitzer, Chief Medical Officer of Aetna, join Dr. Shiv Rao on stage to discuss AI-enabled value-based care. 00:49:29 — NVIDIA Partnership: Clinical Conversation Foundation Model Kimberly Powell, VP of Healthcare at NVIDIA, joins Dr. Shiv Rao to announce the two companies are developing the first foundation model purpose-built for the clinical conversation, on top of NVIDIA's Nemotron model family. Powell walks through the "why now": three massive AI breakthroughs in 18 months (generate → reason → do work), and makes the case that generic AI simply cannot handle the complexity of healthcare workflows.