AI

HLTH 2026 opens in Las Vegas with healthcare AI under pressure to prove it can scale

HLTH USA 2026 opened at the Venetian Expo in Las Vegas with artificial intelligence dominating the program, alongside questions of clinical validation, interoperability, costs and scale.

Author
HDM Staff
Save
Share
Blue and green lights shine over the audience and main stage at HLTH USA 2026.
Photo credit: HLTH USA 2026

    At the Venetian Expo, a four-day program spanning AI, clinical care, data infrastructure and healthcare finance puts implementation and evidence alongside innovation.

    LAS VEGAS — HLTH USA 2026 opened Sunday at the Venetian Expo with artificial intelligence occupying an unusually prominent position in the healthcare industry’s technology conversation. The meeting, which continues through Wednesday, brings vendors, clinicians, insurers, investors and health-system executives together at a moment when the question is increasingly not whether AI can produce an impressive demonstration, but whether a healthcare organization can deploy it dependably across clinical and administrative work.

    The conference’s own program overview shows a packed opening day of payer, clinician and industry sessions before the Sunday main-stage program and welcome reception. On Monday, the exhibition floor opened for a full day as the agenda moved into AI-focused sessions, business meetings and broader discussions about how technology might change care delivery. Photographs provided to Health Data Management show the illuminated HLTH entrance, a busy exhibition hall and elaborate exhibitor displays at the Las Vegas venue.

    HLTH promotes attendance of more than 12,250 people and participation from more than 950 sponsors and exhibitors. Those figures come from the organizer’s event website rather than an independent attendance count. The significance for healthcare executives lies less in the size of the gathering than in the diversity of buyers, technology companies and clinical stakeholders being asked to make consequential decisions about the same emerging tools.

    From AI pilots to everyday care

    The center of gravity is the dedicated AI Zone, which HLTH describes as a venue for short presentations, demonstrations and discussions of deployed technology. Its stated topics extend from ambient clinical documentation and generative AI to predictive analytics, diagnostic imaging, workforce effects, responsible AI, patient-facing applications and the interoperability infrastructure on which many of those products depend. The range matters: the same two-letter abbreviation now covers products with very different intended uses, risk profiles and evidentiary needs.

    An important backdrop is that AI use among physicians is already widespread, even if its clinical impact remains uneven. In a March survey, the American Medical Association reported that 81% of 1,692 participating physicians used AI professionally, more than double the 38% reported in 2023. The largest reported uses were research and standards-of-care summaries, drafts of clinical documentation and administrative support. Use for assistive diagnosis was far less common, illustrating why a high adoption statistic should not be mistaken for proof that automated clinical decisions have become routine.

    Physician confidence is also conditional. In the same AMA findings, 88% of respondents identified validation of safety and effectiveness as important to wider use, and 86% emphasized data-privacy assurances. Those concerns frame a practical challenge for the companies demonstrating products in Las Vegas: executives need to know what happens after a system is integrated with an electronic health record, who reviews its output, how errors are detected and whether an efficiency gain holds up under real-world staffing pressures.

    The market is moving faster than a single regulatory category

    The event’s speaker lineup includes Sumbul Desai, Apple’s vice president of Health and Fitness; Ashley Alexander, OpenAI’s vice president of product for health; and Syed Mohiuddin, Anthropic’s head of healthcare. Their scheduled participation reflects the interest of large consumer-technology and foundation-model companies in the sector. It does not, by itself, establish that any particular product was launched or that a clinical result has been independently verified at HLTH.

    For products that fall under its device authorities, the Food and Drug Administration has an expanding regulatory record. The agency said in its September 2026 update that it had authorized more than 1,600 AI-enabled medical devices for U.S. marketing. That is an important measure of activity, but it is not a count of all healthcare AI tools: many documentation, scheduling, research and administrative products fall outside medical-device regulation, while even authorized devices require attention to their specific intended uses and operating conditions.

    Regulators have also emphasized the information needed to assess decision-support technologies. Federal health IT certification guidance describes transparency and risk-management requirements for certain predictive decision-support interventions supplied as part of certified health IT. Those obligations do not constitute a universal approval system for all AI used in a hospital. They do, however, sharpen the purchasing question: can customers obtain enough information about a model’s purpose, evidence, limitations and risk controls to use it responsibly?

    AI is only part of the cost and care-delivery debate

    The official HLTH agenda circuits extend well beyond generative AI. Thirteen themed tracks address the business of healthcare, behavioral health, consumer health, employer needs, GLP-1 medicines and cardiometabolic disease, aging, oncology, life sciences, specialty care, women’s health and startup finance, among other subjects. This breadth reflects a system in which technology acquisition is inseparable from reimbursement, staffing, affordability and access.

    For payers and providers, technology promises often converge on data sharing and operational workflow. A tool that predicts a patient’s risk, produces a visit summary or recommends follow-up care may be technically strong and still fail if information cannot move between organizations, the clinical team cannot act on the recommendation, or the cost of implementation exceeds the measurable benefit. The strongest evidence of value is therefore likely to come from completed deployments, not product demonstrations or prospective projections.

    HLTH’s daily schedule includes formal investor and market-connection meetings alongside main-stage programming and the exhibition floor. That structure gives the gathering a commercial function as well as an editorial one: technology developers are seeking customers and financing while healthcare organizations weigh their next purchasing decisions. Vendor claims offered in that setting warrant the same scrutiny as any other commercial evidence, particularly when clinical safety, patient data or systemwide costs are involved.

    What to watch as the conference continues

    The most consequential announcements through Wednesday will be those that answer specific operational questions. Has a product been tested with real patient populations? Is it in general release or only a pilot? How does it connect with existing information systems? What is the independent evidence on care quality, staff workload or cost, and who bears responsibility when the technology fails? Those distinctions are essential to separating industry momentum from measurable progress.

    The Las Vegas show floor makes the scale of healthcare’s technology market visible. Whether HLTH 2026 marks a genuine shift toward more reliable, affordable and accountable care will be determined away from the conference lighting — in health systems and clinics that must decide which capabilities are ready, which need more evidence and which should not yet be deployed.

    Photography captured by Health Data Management staff at HLTH USA 2026.