The ASTP Annual Meeting took place February 11-12, 2026, in Washington, DC, bringing together health IT leaders, policymakers, clinicians, and industry representatives for two days of sessions focused on the future of health information technology. ASTP (Assistant Secretary for Technology Policy), formerly known as the Office of the National Coordinator for Health IT (ONC), is the federal body charged with advancing health IT adoption and nationwide health data exchange.
The overarching theme of the meeting was a sense that interoperability tools (e.g., TEFCA, FHIR, USCDI) are mature enough that the healthcare community should be able to deliver on the long-promised vision of seamless health data exchange. TEFCA was a recurring topic across many of the sessions, with a lot of high-level enthusiasm and advocacy; though some discussions moved past the basics and into the harder questions of quality over quantity of data, rural access, compliance enforcement, and technical performance.
The fireside chat between Tom Keane and Amy Gleason captured the administration’s ambitions, from e-prior authorizations and provider directories to a future where patients can consolidate all their health data in a single app and use AI to reduce unnecessary doctor visits and shop on price. This and other sessions reinforced that patient empowerment is a policy priority, but it was also clear that the payer-patient friction holding back much of the empowerment vision is well entrenched and will be slow to change.
AI was everywhere, and the conversation was notably optimistic and at times starry-eyed. We heard one panelist actively discourage publishing negative AI studies on the grounds that they are immediately obsolete, which is worth noting as a reflection of the current mood. A plug for the HealthEx/Claude deal and a presentation of DocTronic’s AI-prescription model illustrated how quickly agentic AI is moving from concept to product. Panelists talked about a vision of AI “doctors” being supervised by other AI Agents so that real doctors can practice at the top of their license. We heard the Food and Drug Administration (FDA) talk about flexibility and collaboration with industry and how regulators need to accommodate the fast pace of AI. It’s interesting to think about how that AI-friendly stance may impact Clinical Decision Support guidance: If a CDS product (such as HLN’s Immunization Calculation Engine) is not considered a regulated medical device because it provides recommendations rather than specific directives; but the clinician who interprets its output and produces directives is an AI Agent, how should that be regulated?
The data liquidity and API sessions grounded things a bit, making clear that the infrastructure that we are working to build on the ground (with bulk FHIR, bidirectional APIs, CDS Hooks, etc.) is still catching up to the vision. A point made in a session on emergency services – that “log in to the portal” is not an acceptable solution in a health crisis – was one of the more memorable moments of the event.
There was a notable lack of content related to public health, with no public health-oriented sessions and no panelists from CDC or from State, Tribal, Local, and Territorial Public Health agencies. But the HTI-5 comment period and the AI RFI are opportunities for public health to provide feedback from its unique perspective on these initiatives.
Overall, across both days the meeting reflected a strong federal emphasis on interoperability, deregulation, price transparency, and harnessing AI to improve health outcomes and reduce provider burden.