Seattle, March 6, 2026, 4:32 AM UTC-08:00
Amazon.com’s cloud arm rolled out a new AI tool for healthcare on Thursday, stepping further into a sector where hospitals are eager to automate phone calls, paperwork, and billing. The new offering, Amazon Connect Health, taps what AWS is calling agentic AI — software built to handle tasks with minimal human input — and integrates directly into electronic health records, or EHRs. That lets it check patient details, handle appointment scheduling, draft notes, and produce billing codes, all within the digital systems clinicians already use.
Timing is key here. Investors want to know when Amazon’s big bets on AI will turn into more obvious gains. Last month, Reuters said Amazon is eyeing roughly $200 billion in capital spending for 2026. D.A. Davidson’s Gil Luria didn’t mince words: the company “has to invest at these levels just to stay in the race.” Reuters
The race in healthcare tech isn’t slowing down. On Thursday, CVS Health and Google Cloud announced plans to roll out an AI-driven health platform. Microsoft has taken a different tack, leaning on Nuance—acquired in 2021, as Reuters reported—to boost its suite of cloud-based provider tools and clinical documentation.
AWS has rolled out patient verification and ambient documentation—software that records visits and generates notes—as generally available features. Appointment management, patient insights, and medical coding are still in preview. Naji Shafi, AWS’s director of healthcare AI, told MobiHealthNews the tech ties directly into Epic for real-time patient verification and appointment handling.
AWS wants to frame the product as more than just another feature—it’s aiming for a comprehensive workflow tool. Rajiv Chopra, vice president of Health AI and Life Sciences at AWS, told GeekWire the company isn’t chasing “point solutions.” An AWS blog pointed out that in large health systems, staffers can spend as much as 80% of their call-handling time piecing together patient data from various unconnected systems. GeekWire
Amazon cited fresh customer data to make its argument. UC San Diego Health, responsible for 3.2 million patient interactions annually, saved about a minute per call, according to Amazon. The health system managed to redirect 630 staff hours each week away from verification and toward patient support, while average call abandonment dropped 30%—hitting as much as 60% in some departments. Amazon added that One Medical has now used the documentation tool in over a million visits.
AWS claims its service runs around the clock, flags more complex cases for human review, and ties each piece of AI-generated documentation or code to its origin via what it describes as evidence mapping. For clinicians, that translates to being able to access the full underlying transcript or record that produced the output, rather than having to rely purely on the software’s word.
Still, healthcare AI isn’t an easy pitch. Back in January, Reuters flagged that these ambient scribe tools are already triggering concerns about privacy, malpractice, and consent—especially as clinics scramble to adopt them and navigate current state and federal regulations.
The accuracy problem is still hanging over AI. Back in February, Reuters highlighted a study in The Lancet Digital Health showing that AI models are likelier to pick up and repeat medical errors found in documents that sound credible. “Current AI systems can treat confident medical language as true by default,” said Dr. Eyal Klang of Mount Sinai. Reuters
Uncertainty is now tangled up with Amazon’s bigger spending gamble. Russ Mould, investment director at AJ Bell, pointed out that the stock’s drop after Amazon’s February spending announcement signaled a shift—investors steering clear of names “where positive surprises may be hard to achieve.” As a result, launches like Connect Health are drawing sharper scrutiny compared to a typical AWS rollout. Reuters
