
Sign up to save your podcasts
Or


A high score on the USMLE does not prove that an AI system is ready for use in the clinic. Clinical AI needs to be able to handle changing patient information, recognize missing or conflicting context, ask useful follow-up questions, and support care through real clinical workflows.
In this episode of Between Care and Code, Spencer Wozniak and Luis Cisneros, co-founders of Serelora, discuss why medical exam benchmarks don’t capture the full complexity of patient care. They explore “context readiness,” a proposed way to evaluate whether clinical AI has enough relevant information to support a decision, and whether it can identify what information is still needed.
The conversation covers AI in healthcare, clinical AI benchmarks, USMLE scores, EHR data, AI agents, patient context, and human oversight. The goal isn’t just to test whether AI can answer a medical question, but whether it can help clinicians move care forward safely.
Read the full article here: https://www.thelancet.com/journals/landig/article/PIIS2589-7500(26)00147-0/fulltext
0:00 Introduction to Context Readiness in AI
02:38 Evaluating Clinical AI Benchmarks
05:09 The Role of Data in AI Models
08:53 Benchmarking AI Models in Clinical Settings
11:04 The Importance of Context in AI Decision Making
12:53 Integrating Clinical and Billing Workflows
25:27 Interoperability vs. Portability in Healthcare
Can AI agents safely and legally work inside electronic health record (EHR) systems? Spencer Wozniak and Luis Cisneros discuss the Heidi Health v. Best Practice dispute, healthcare data access, interoperability, and the role of legacy EHRs.
They also dig into AI accountability: who should review an agent’s work, how audit trails can support human oversight, and whether AI can ease clinicians’ administrative burden without losing sight of patient care. A conversation about AI in healthcare, EHR interoperability, clinical workflows, and the future of health tech.
Keywords: AI in healthcare, AI agents, EHR, electronic health records, healthcare interoperability, clinical AI, patient data access, AI governance, health tech.
Can AI agents finally break down the barriers that have kept healthcare data fragmented for decades?
In this episode of Between Care and Code, Serelora co-founders Luis Cisneros and Spencer Wozniak explore how artificial intelligence is challenging traditional electronic health records (EHRs), healthcare interoperability, and the administrative systems that shape modern medicine.
From patient trust in clinical AI to expensive EHR integrations, insurance claims, medical record access, and autonomous AI agents, we examine what happens when technology begins to expose the inefficiencies built into healthcare.
We also discuss human oversight in clinical decision-making, open-source AI models, quantitative risk stratification, and why the next generation of healthcare technology should focus on improving the practice of medicine rather than simply automating existing bureaucracy.
Topics include
- AI agents and the future of electronic health records
- Healthcare interoperability, FHIR APIs, and patient data access
- Human-in-the-loop AI and clinical decision-making
- Automating insurance claims and administrative workflows
- Open-source AI versus proprietary models
- Clinical intelligence, risk stratification, and longitudinal patient data
- How Serelora is rethinking the EHR as a system of record and intelligence
Between Care and Code explores the intersection of medicine, artificial intelligence, healthcare technology, and the systems shaping the future of patient care.
Learn more at https://serelora.com
Serelora co-founders Spencer Wozniak and Luis Cisneros discuss what trustworthy clinical AI should look like, prompted by a dispute involving OpenEvidence and Doximity. Drawing on clinical experience and product-building lessons, they argue for transparent reasoning, traceable sources, physician oversight, and AI that augments rather than replaces medical judgment.
Keywords
clinical AI, healthcare AI, medical AI transparency, explainable AI, clinical decision support, AI scribes, physician liability, evidence-based medicine
Key Topics
Takeaways