Daniel Vreeman, DPT, Chief Standards Development Officer, HL7 International
LinkedIn: Daniel Vreeman
AI has moved quickly from buzzword to boardroom priority in healthcare, yet many organizations are still struggling to translate early experimentation into measurable, scalable results. At the same time, demand is growing for AI that is trustworthy, explainable, and designed to flourish within real clinical and operational workflows, not just in controlled pilots.
As HL7 International launches its second annual Global AI Challenge, a worldwide competition spotlighting solutions that use standards-based data to solve real healthcare problems, we spoke with HL7’s Chief AI Officer, Dr. Daniel Vreeman about what last year’s inaugural competition revealed, where healthcare AI is headed next, and why interoperability is a foundation for innovation. For full competition details and submission guidelines, visit, https://info.hl7.org/ai-challenge. Submissions are open through June 30, 2026.
Q: Why did HL7 create an AI Challenge in the first place?
Dr. Vreeman: We saw a gap in the industry narrative about AI. There were plenty of conversations about AI models, but fewer examples of solutions grounded in the realities of healthcare data, interoperability, and workflow integration. HL7 is the place where the industry gathers to build healthcare’s interoperable future together. We wanted to create a forum where innovators could demonstrate how AI can solve real healthcare problems when built on trusted, standards-based data.
Q: What surprised you most from last year’s inaugural competition?
Dr. Vreeman: The creativity and practicality of the submissions. Many entrants were focused not on abstract AI concepts, but on real-world friction points: navigation, access, communication, clinician burden, and patient understanding. That was encouraging because it reflects a maturing market. Healthcare organizations are increasingly asking not ‘Can we use AI?’ but ‘Where does AI actually improve outcomes or efficiency?’
Q: What does the second year of the challenge say about where healthcare AI is today?
Dr. Vreeman: It tells us we’re primed for a new phase. The first wave of healthcare AI was largely about pilots and proofs of concept. The next phase is about scale, trust, and integration. Health systems want AI that works within existing workflows, uses reliable data, can be governed responsibly, and delivers measurable value. That’s a significantly higher bar than simply demonstrating technical capability.
Q: You’ve said AI is only as strong as the data behind it. What does that mean in practice?
Dr. Vreeman: AI depends on accurate, timely, and well-structured information. In healthcare, data often comes from many systems like EHRs, devices, payers, labs, patient apps, etc., and those systems don’t always speak the same language. That’s where standards matter. Standards like FHIR help create consistent ways to exchange and understand data, which is foundational if you want AI outputs to be trustworthy and actionable.
Q: What kinds of AI use cases seem most promising right now?
Dr. Vreeman: We’re seeing strong momentum in areas like administrative efficiency, documentation support, patient engagement, care navigation, and surfacing insights from existing data. These are important because they can create near-term value while organizations continue building the infrastructure needed for more advanced clinical applications.
Q: What still needs to be solved?
Dr. Vreeman: Governance is a major issue, things like explainability, transparency, monitoring, bias mitigation, and lifecycle management. We also need to make adoption easier, especially for smaller and less resourced delivery organizations. Even great tools can fail if they create workflow disruption or require heroic effort to implement.
Q: What do you hope to see from this year’s challenge?
Dr. Vreeman: Solutions that combine innovation with readiness. We want to see ideas that are technically strong, but also practical, responsible, and scalable. The most exciting AI solutions in healthcare may be the flashiest. Often it’s the solution that quietly fits into care delivery and makes things work better for clinicians and patients.
Q: What message would you send to health systems watching the AI market right now?
Dr. Vreeman: Focus less on hype and more on foundations. Ask whether your data is ready, whether workflows are ready, and whether you have scalable, dynamic governance in place. AI success in healthcare will depend less on who adopts first, and more on who builds responsibly.