AI in Action
Artificial intelligence has shifted from an experiment to an everyday operational force in healthcare. During the recent American Hospital Association’s Leadership Scan, supported by CorroHealth and Prenosis,, executives from Cedars-Sinai, Houston Methodist, Sanford Health, and CorroHealth described how AI is changing workforce dynamics in real and measurable ways.
The discussion, led by AHA Vice President Lindsay Dunn Bergstahler, centered on one question: how can AI ease administrative pressures and return clinicians’ focus to patient care? Their answers showed that AI’s value lies less in its distant promises and more in its tangible ability to reduce fatigue, eliminate inefficiency, and enhance satisfaction across the healthcare workforce.
A Clearer View of AI’s Role
Tammy Knobbe, Vice President at CorroHealth, opened with a reminder that not all AI is created equal. Automation handles repetitive tasks with minimal human oversight while artificial intelligence applies human-like reasoning. Generative AI, the newest widely adopted form of AI, produces new content or text from large language models. Agentic AI goes further, making goal-directed decisions autonomously.
Understanding these distinctions helps organizations plan responsible adoption strategies. For many hospitals, automation and generative tools have already proved their worth. Others are beginning to experiment with agentic systems that can, under human supervision, make real-time scheduling, routing, and workflow decisions.
Reducing Clinician Burnout
At Sanford Health, Chief Medical Officer for Virtual Care David Newman described how ambient listening technology transformed the workday for hundreds of physicians. The system listens during patient visits, transcribes conversations, and automatically generates documentation for review. In only a year, nearly 300 physicians adopted the tool. Every single one said they would be disappointed to lose it.
Newman explained that 88 percent of participating clinicians reported reduced burnout and fatigue; 90 percent said their job satisfaction improved; and 95 percent felt their cognitive load had decreased. The technology, once seen as optional, is now essential. Recruiting new physicians increasingly depends on providing this kind of support. Newman noted that the investment is substantial, but the return is clear: satisfied clinicians deliver better care and are far more likely to stay.
Nursing Relief at Cedars-Sinai
Cedars-Sinai Medical Center’s Executive Director of Nursing, Peachy Hain, offered an equally powerful example from the front lines of care. Her team launched Ava Nurse Assistant, a mobile voice application that lets nurses record care activities directly from their phones. Instead of walking to a workstation and entering data manually, nurses speak naturally and review what the AI has transcribed before approving it for the medical record.
The shift produced immediate impact. Nurses no longer delay documentation until the end of a shift, cutting incidental overtime nearly in half. Patient experience scores rose by more than thirty points, and staff satisfaction followed. Hain said the technology freed nurses to “focus on care instead of keyboards.”
Virtual nursing has further eased the burden by letting remote nurses handle admissions and screenings, allowing bedside teams to concentrate on direct patient care. For Cedars-Sinai, AI has become less about automation and more about restoring human connection.
Operational Gains at Houston Methodist
At Houston Methodist, Executive Vice President and Chief Innovation Officer Roberta Schwartz described AI as “embedded in almost everything we do.” Predictive analytics now guide scheduling and patient flow, helping the organization increase surgical throughput by roughly fifteen percent during peak hours. AI tools analyze vital sign data in real time, prompting timely interventions when patient conditions start to deteriorate.
Schwartz cautioned, however, that agentic AI is not always the right answer. Some implementations add cost or complexity instead of reducing it. Her team focuses on use cases with clear efficiency gains and on maintaining rigorous evaluation after each rollout. “You can’t adopt everything at once,” she said. “Innovation requires choosing carefully where to invest attention and change capacity.”
Streamlining the Business of Healthcare
Knobbe explained that AI’s impact extends beyond the clinical setting. Generative technology now help hospitals prepare documentation for payer appeals, identify coding inconsistencies, and spot documentation gaps before they trigger denials. She noted that intelligent technology–especially clinical GenAI trained on large volumes of validated medical records and coding criteria–can identify opportunities for more accurate, documentation-supported diagnoses with far greater consistency than manual review.
By using context-aware systems rather than simple automation, organizations have reduced unnecessary physician queries and accelerated revenue cycle workflows. Early results show success rates of up to seventy percent in identifying improvement opportunities, which is significant progress for functions once driven entirely by manual review.
Knobbe added that these post-discharge, pre-bill applications help hospitals address challenges that frontline tools cannot reach: reducing query fatigue, supporting stronger DRG validation, and mitigating denials before they impact revenue. While much of the panel focused on in-the-room clinical support, she emphasized that easing administrative pressure after the patient encounter is equally critical to workforce stability.
Managing Fear and Building Trust
Introducing AI into healthcare inevitably raises questions. Staff often worry that new tools will replace their jobs or monitor their conversations. Hain said the best way to manage these fears is through transparency. From the first day of implementation, Cedars-Sinai explains what each system can and cannot do. Patients and nurses are reassured that voice tools record only clinical documentation, rather than private conversations.
Newman reframed the discussion to focus on partnership rather than replacement. In his specialty, endocrinology, he pointed out that an “artificial pancreas”—a closed-loop insulin delivery system—already outperforms human monitoring. “Sometimes AI is better than the doctor,” he said, “and that’s okay. It’s our job to partner with it responsibly.”
Kenobi reminded listeners that AI reflects the biases of its training data. Organizations must establish governance models and maintain human oversight to prevent algorithmic bias from affecting care decisions.
Measuring Value and Scaling Responsibly
Each speaker agreed that clear metrics drive successful adoption. Cedars-Sinai evaluates new tools against three criteria: impact on patient care, effect on clinical workflow, and adherence to safety and ethical standards. The Ava pilot succeeded on all counts, leading to a full rollout.
Houston Methodist relies on continuous ROI tracking. Schwartz compared innovation management to cleaning a closet: if a tool hasn’t proven its value recently, it’s time to retire it. At Sanford, Newman stressed the importance of validation through evidence, rather than vendor claims, especially when patient safety is involved.
Preparing Today’s Workforce for Tomorrow’s Tools
The arrival of AI has changed workforce planning as much as day-to-day operations. Sanford Health has redeployed administrative staff from scheduling roles to higher-value clinical monitoring positions. Houston Methodist has used AI to reduce contract labor and adjust staffing ratios in select units without sacrificing care quality.
Kenobi added that administrative skill sets are evolving as automation expands. Tomorrow’s medical coders, she said, will need to understand data logic and validation as much as diagnostic codes. For Hain, the shift is even more fundamental. “AI is essential to help clinicians manage their growing workloads,” she said. “It allows us to spend time with patients again.”
Looking Ahead
When asked to predict what comes next, the panelists shared cautious optimism. Newman expects AI capabilities to become a core expectation for incoming physicians. Schwartz foresees diagnostic breakthroughs from imaging and pathology models capable of detecting dozens of conditions from a single scan. Kenobi anticipates new governance structures to manage rapid technological change. And Hain envisions a nursing future where every caregiver has a digital assistant capable of tracking reminders, synthesizing data, and supporting decisions in real time.
Across all perspectives, one theme remained constant: AI’s true promise lies in making healthcare more human, not less. By removing the friction of administrative work, technology can return time, attention, and empathy to the center of care.
As Hain put it, “We need AI to take care of the clerical work so we can take care of our patients.”
Knobbe reflected that the same principle applies beyond bedside workflows. When AI improves coding accuracy, strengthens documentation, and reduces compliance risk, it lowers pressure on clinicians and revenue cycle teams alike. The cumulative impact (clinical and administrative) creates the conditions for a more stable workforce.
Next Steps
Discover how CorroHealth partners with healthcare leaders to harness AI, improving care quality and reducing staff burden.