Mass General Brigham has deployed an AI intake agent inside MyChart that helps route primary-care patients to the right appointment type, with the health system reporting that 77% of patients using the tool get seen the same day they reach out. The figure, reported October 8 by OpenLoop Health, offers one of the more concrete productivity metrics yet published for a hospital-system AI deployment.
How the Intake Agent Works
Rather than requiring patients to navigate a phone tree or manually select an appointment type when they report new symptoms through MyChart, the AI agent interprets the patient’s described concern and routes them toward the appropriate care pathway — urgent same-day slots, standard primary-care scheduling, or a nurse triage line — based on patterns the system has learned from historical scheduling and outcome data. That automated triage function has traditionally required a human scheduler or nurse to perform manually, a bottleneck that has contributed to primary-care access delays across the US healthcare system.
Why Same-Day Access Matters So Much
Primary-care access delays have been directly linked in health-services research to higher downstream emergency-department utilization, since patients who can’t get a timely appointment for a new symptom often end up in urgent care or the ER instead, at significantly higher cost to both the patient and the health system. Mass General Brigham’s reported 77% same-day rate, if sustained at scale, would represent a meaningful improvement in a metric health systems have struggled to move despite years of scheduling-software and staffing investments predating the current AI wave.
What the Reported Figure Doesn’t Tell Us
The 77% statistic, drawn from OpenLoop Health’s reporting, doesn’t specify the total patient volume the AI agent has processed, nor does it break down performance by clinic location, patient demographics, or symptom type — details that would help assess whether the tool performs consistently or whether its strong headline number reflects a subset of simpler, easily triaged cases. Health-services researchers typically caution against drawing firm conclusions from single aggregate statistics without that kind of stratified breakdown.
The Broader Trend Toward AI Patient Navigation
Mass General Brigham joins a growing list of large health systems deploying AI-driven patient-navigation tools, part of a broader shift in how hospitals are using generative AI — not primarily for diagnosis, where regulatory and liability concerns remain significant, but for operational tasks like scheduling, triage routing and administrative documentation, where the stakes of an imperfect AI suggestion are lower and the labor-cost savings are more immediate and easier to quantify.
What Clinicians and Patients Have Said
Physician reaction has generally focused on reduced administrative burden for front-desk and triage staff, freeing clinical personnel to focus on direct patient care rather than scheduling logistics. Patient feedback specific to the MyChart AI agent has not been separately published, leaving an open question about how patients themselves experience interacting with an AI system for what has traditionally been a human-staffed scheduling conversation.
What’s Next
Mass General Brigham has not announced plans to expand the AI intake agent beyond primary care into specialty scheduling, though the same underlying triage logic could plausibly extend to other service lines if the primary-care results hold up under more detailed scrutiny from health-services researchers and the health system’s own quality-improvement teams.
What Health-Services Researchers Want to See Next
Researchers who study care-access interventions caution that a single aggregate statistic, however impressive, rarely tells the full story of how a new triage system performs across the full diversity of a health system’s patient population, and they typically want to see the 77% same-day figure broken down by clinic location, insurance type, primary language and presenting complaint before drawing strong conclusions about the tool’s overall effectiveness. Mass General Brigham has not yet published that kind of stratified data, leaving outside researchers to rely on the health system’s own framing of the result for now. There is also a question of substitution effects: health-services economists want to know whether the AI agent is genuinely creating new same-day capacity through smarter routing, or whether it is simply redistributing a fixed pool of same-day slots more efficiently among patients who request them, a distinction that matters enormously for whether the tool represents a genuine capacity expansion or primarily a better allocation of existing capacity. Other large health systems watching Mass General Brigham’s rollout, including several that have piloted similar AI intake tools on a smaller scale, have told trade publications they are waiting for exactly this kind of more granular outcome data before committing to system-wide deployment, reflecting a broader pattern in which health-system AI adoption decisions increasingly hinge on demands for more rigorous, stratified evidence than vendors have historically needed to provide to close a sale.
Photo: HUUM │sauna heaters / PEXELS via Pexels
