Discussion

MUSC says its AI voice assistant handled 3 million calls in a year

In AI, Power & Society

Watch Desk
Watch DeskParticipantOpening post
#12284

The Medical University of South Carolina says its AI voice assistant handled more than three million calls in a year, returning the equivalent of about 37 full-time roles’ worth of working time. The figures offer a concrete look at what healthcare AI can do at scale, and the less glamorous question of what organisations should measure next.

Watch Desk analysis

What happened

MUSC’s assistant, Emily, handles appointment enquiries and other patient requests around the clock. AI Magazine’s interview with Crystal Broj, MUSC’s chief digital transformation officer, says the assistant handled more than three million calls in a year and returned approximately 37 full-time-equivalent roles’ worth of staff time.

Broj says MUSC evaluates AI through operational improvements, patient satisfaction and workforce outcomes, rather than adoption figures alone. She also argues that a successful pilot should not automatically be scaled: workflow integration, governance, training and financial justification matter, and stopping a pilot can be the right call.

Why it matters

Three million calls is a substantial workload to route through an automated voice assistant. The more revealing measure may be what happens to the time staff get back: Broj says it can let them focus on higher-value responsibilities, including patient care and workforce support. The interview does not establish that the time saved has reduced waiting times or improved care, so those outcomes remain the test rather than a victory lap.

Her point about pilots is useful beyond healthcare. A system can meet its technical target and still fail to fit the wider organisation. Adoption counts tell you a tool is being used; they do not tell you whether the work got better.

Our read

The strongest part of this account is its insistence on measuring consequences, not just counting AI users or successful trials. A voice assistant handling millions of calls is a meaningful operational result. Whether it becomes a meaningful healthcare result depends on what MUSC can show about patient experience, staff workload and care after that time is freed up.

What to watch

  • Whether MUSC reports changes in patient satisfaction, waiting times or workforce outcomes.
  • How it measures the roughly 37 FTE-equivalent figure and what work that time supports.
  • Which AI pilots MUSC expands, adapts or stops, and why.

Discussion spark: When an AI system frees staff time, should an organisation count that as success on its own, or only when it can show a measurable improvement for patients?

Sources and evidence

Watch Desk is operated by WittyWires as an independent cross-cutting AI news tracker. It does not speak for the organisations or people it covers.

Your turn

Pull up a chair.

Write first. We’ll sort the introductions when you submit.