UI Health is rolling out an AI documentation system across its inpatient, outpatient, and emergency-care settings. The academic health enterprise of the University of Illinois Chicago treats a multilingual patient population of about 1.4 million people each year. The system selected Abridge, a Pittsburgh-based company that builds automated clinical documentation tools designed to convert clinician-patient conversations into structured medical notes.
UI Health said it prioritized tools that improve the experience of both clinicians and patients. “We prioritize technology that enables the best experience for both our workforce delivering care and the patients receiving it,” said Dr. David Chestek, Chief Medical Information Officer at UI Health.
The decision followed a pilot in which a key patient-experience measure improved. Survey responses to “provider explained in a way you understand” increased from 91 to 97 percent during the test period. The health system has presented the deployment as part of an effort to reduce documentation burden and improve clarity in clinical encounters.
Abridge’s system integrates directly into Epic, UI Health’s electronic health-record platform, allowing clinicians and trainees to generate AI-assisted documentation without leaving their existing workflow. Abridge positions its platform as enterprise-grade and says it supports more than 28 languages and more than 50 medical specialties. These capability claims come from the company’s own materials.
Abridge said UI Health selected the system following a competitive evaluation, though the institution has not publicly described the procurement process. The company’s description of the selection process appears in its own blog post.
Shiv Rao, Abridge’s CEO and co-founder, said the goal of the deployment is to improve the way clinicians communicate with patients. “Abridge helps reflect one of the things that matter most: enabling clinicians to meet their patients where they are,” Rao said.
The platform uses speech recognition trained to support clinical conversations in English and multiple additional languages, including Spanish, Arabic, Polish, Mandarin, and Cantonese. UI Health serves communities where many patients do not speak English as their primary language.
The institution’s public statements frame communication as a central motivation for the rollout. Its leadership has described the tool as a way to ease documentation responsibilities while supporting patients who may require clinicians to shift between languages during visits .
Adapting for a multilingual population
UI Health’s patient base reflects the demographics of Chicago’s West and South Sides, where Spanish, Arabic, Polish, and Chinese dialects are common. The enterprise includes a tertiary-care hospital, outpatient clinics, and community health centers. It has emphasized that many patients communicate in multiple languages or rely on interpreters, creating challenges for clinicians attempting to capture accurate, timely notes.
Abridge’s tool records conversations, produces drafts of clinical notes, and incorporates context from prior documentation. The pilot results at UI Health show that the system may help translate spoken conversations into clearer documentation for patients. The improvement from 91 to 97 percent in the communication measure is the main outcome the institution has disclosed.
Other health systems have cited similar reasons when adopting Abridge. AltaMed Health Services Corporation, the nation’s largest federally qualified health-center network, announced a deployment earlier this year and said the decision aligned with its need to serve multilingual patient populations. Cambridge Health Alliance, a public health system in Massachusetts, also selected Abridge and emphasized the platform’s multilingual capabilities as a core reason.
Independent literature has also begun to examine the impact of ambient clinical documentation tools. A peer-reviewed evaluation of Nuance’s DAX platform, a competing product, reported positive provider-engagement trends without compromising documentation safety or patient-experience measures. These findings do not evaluate Abridge but show that large U.S. systems have begun studying AI-driven documentation in clinical settings.
A systemwide rollout
The Abridge deployment embeds AI-generated documentation directly into UI Health’s Epic interface. This makes the tool part of routine care rather than an additional application that clinicians must adopt. Integrating ambient documentation within Epic matters in academic medical centers where hundreds of residents, fellows, and faculty use shared workflows.
UI Health’s approach aligns the tool with existing systems for quality, compliance, and billing, which reduces the operational friction that has slowed AI adoption elsewhere. The rollout spans inpatient units, outpatient clinics, and emergency departments, and applies to both attending physicians and trainees. UI Health described the implementation as enterprise-wide.
The institution did not release a detailed timeline or specify whether individual departments will be onboarded simultaneously or in phases. Its public statements focus on documentation burden, communication quality, and the goal of standardizing workflows across care settings.
Abridge says its platform is used by more than 200 large and complex health systems. Several of these, including AltaMed and Cambridge Health Alliance, have confirmed deployments (AltaMed, CHA). UI Health’s decision places it among a group of systems that are adopting AI documentation not only for operational efficiency but also for language support and patient-experience outcomes.
The institution has not disclosed long-term performance data beyond the pilot’s communication metric, and no independent audits of Abridge’s multilingual capabilities have been published. For now, the rollout marks one of the more comprehensive ambient documentation deployments in a U.S. academic medical center.
