James Bates, founder and CEO of Abby Medical, joins Sunny Ray to explain how artificial intelligence can automate almost every part of running a medical practice, leaving only the medical decision itself to the physician. Bates traces his path from a rebellious car-obsessed kid in Southern California to an electrical engineer at Bell Labs working on the first digital cell phone chipset, then through leadership roles at Silicon Labs, Maxim, and Freescale building networking, IoT, and self-driving vehicle technology. Living in Japan exposed him to a single-payer healthcare system riddled with administrative waste, which shaped his belief that inefficiency, not payment models, is healthcare's core problem. After selling his last company, he applied engineering discipline, including a Six Sigma lean analysis, to medical practices and discovered doctors spend most of their time on billing and liability rather than patient care. His company Abby Medical uses an AI assistant named Hannah to gather patient information, predict diagnoses and insurance outcomes, and let physicians focus purely on medicine again.
A veteran chip and self-driving tech executive explains how AI can automate medical practices so doctors can finally practice medicine again.
What is Abby Medical?
Abby Medical uses artificial intelligence to completely automate the medical practice. The AI engages with the patient before the visit, predicts every question the doctor will ask, predicts the treatment and whether insurance will pay, and gives the physician a one-page summary. Once the physician confirms the diagnosis, the AI scribes the encounter, codes it, and sends it out. Everything is automated except the medical decision itself.
What was your first real job in tech?
That was at Bell Labs. I was a research designer for GPRS, working on the very first digital cell phone chipset standard worldwide. I had the opportunity to work on that straight out of graduate school.
How did you go from engineer to running a billion dollar business?
I was part of the Bell Labs Tokyo design team since I speak Japanese. When that shut down in 1999, I moved into business development, finding lab research and matching it to customers. I later joined Silicon Labs, took the business from zero to half a billion dollars in three years, then ran IoT at Maxim and later led Freescale's self-driving vehicle technology business before selling it and moving into AI and healthcare.
What did you see in other countries' healthcare systems that stayed with you?
In Japan, a single-payer system, care is rationed and doctors are paid per visit rather than per procedure, so something like removing a mole takes five visits over six weeks instead of one. The government ends up paying doctors about the same amount either way, it just forces patients through more visits. I realized every country games the payment system the same way, and the real fix is making care more efficient through technology.
What did the numbers show you when you looked at buying medical practices?
The average profit margin of medical practices in the US is only 3 percent, lower than the Red Cross nonprofit's 5 percent margin from selling blood. I expected high single digits at least, but instead found practices spending huge amounts of time on tasks a machine could do better, which is why margins are so thin.
Why doesn't anyone in healthcare seem incentivized to cut costs?
The Affordable Care Act caps insurance company profit at 20 percent, so insurers maximize dollar profit by growing revenue rather than by becoming efficient. Hospitals face similar dynamics since visible profit invites backlash over pricing. This removes any real incentive to lower costs, making the system function more like a cost-plus organization than a real business.
What does Hannah actually do during a visit?
Hannah is the AI that engages with the patient before the visit, asking every question the doctor would ask, prepping orders, and predicting whether insurance will approve the treatment. That summary lets the physician have a real-time conversation with the patient about diagnosis and options during the actual encounter, cutting out weeks of back and forth with insurance companies.
Building something daring? Sunny talks to founders like this every day. Fifteen minutes to see if your story belongs on the stage.
Claim your pre-interview