Dr. Chadi Nabhan is Chief Medical Officer and Head of Strategy at Ryght AI. The company has built digital twins of every clinical trial site in the world and uses AI agents to match each trial with the sites and investigators most likely to enroll patients. He is a medical oncologist and hematologist who practiced for about 20 years before moving into industry. He joined us in Episode 34 (March 7, 2026).
Before Ryght AI, which he joined about two years before the episode, he was chief medical officer at Cardinal Health and chair of the Precision Oncology Alliance at Caris Life Sciences. He also hosts the podcast Healthcare Unfiltered.
Ryght's goal is to make clinical trials faster, more efficient and cheaper. Dr. Nabhan said a trial can take 10 to 15 years and $1 to $3 billion, and trials fail for three reasons: the drug doesn't work, it is too toxic, or the trial never enrolls enough patients. Ryght doesn't make drugs. It goes after the third reason by finding the right sites and the right principal investigator to champion the trial internally.
80% of trials run behind schedule, 50% of sites enroll just zero to one patients, and from a sponsor perspective, time is money.
— Dr. Chadi Nabhan, Episode 34
Each site's digital twin is linked to its trial history on clinicaltrials.gov. Ryght scores sites and sorts them into tier one, two and three by how likely they are to enroll. He stressed that the best site for a given trial might be number 20 on a generic list.
Dr. Nabhan didn't share his screen. He walked us through how the platform works, starting at 13:21. First, AI infers what a site can do from its past studies. If a site has opened a phase one trial, for example, it can run pharmacokinetic studies. Second, when a sponsor uploads a protocol, 13 agents read it in parallel: one for the title, one for the schedule of events, one for biomarkers, one for endpoints. Within seconds the system knows the trial needs, say, an MRI machine and a CT scanner, and matches that against the digital twins. Third, another agent finds the best contact person at each site.
So when we upload a particular study that could be 200 page study in any therapeutic area, we have 13 AI agents that dissect and parse out, read the entire protocol.
— Dr. Chadi Nabhan, Episode 34
I think having an informed patient and family is 10 times better than having someone who is not informed.
— Dr. Chadi Nabhan, Episode 34
He also has a book on AI and cancer care coming out, written for patients and families rather than technologists. For more on medicine and AI, see AI in healthcare and our conversation with Dov Gertz of Converge Bio, who also expects AI to improve how patients are selected for trials.
Sam built a clinical trial site selection digital twin dashboard with Ryght in mind. It simulates a site network from historical performance data and compares a manual strategy, heavy on tier-one academic centers, with an AI-optimized pick that recovers a two-month delay and hits diversity targets. Toggles for biomarker complexity and a census overlay re-rank the sites. Dr. Nabhan said Sam "certainly picked up on what we're doing" and that ranking sites is exactly Ryght's approach.
Dr. Chadi Nabhan is Chief Medical Officer and Head of Strategy at Ryght AI. He is a medical oncologist and hematologist who practiced for about 20 years, then served as chief medical officer at Cardinal Health and chair of the Precision Oncology Alliance at Caris Life Sciences.
Ryght AI has built digital twins of every clinical trial site in the world that has run a trial, and uses agentic and generative AI to match each trial to the sites and investigators most likely to enroll patients.
Dr. Nabhan says trials fail because the drug doesn't work, it is too toxic, or the trial doesn't enroll. Ryght AI targets the third: 80% of trials run behind schedule and half of sites enroll zero to one patients.
Yes. Ryght uses 13 AI agents, one for the title, one for the schedule of events, one for biomarkers, one for endpoints and so on, to parse a protocol of around 200 pages in seconds and match its requirements to sites.
