AI Use in Health Care 'Barely Scratching the Surface' Says Philips' Jeff DiLullo

AI adoption among health care clinicians is ramping up, with Phillips North America CEO Jeff DiLullo saying the industry has 'barely scratched the surface' of what the technology can achieve. He said AI could improve doctors' quality of work and quality of life by narrowing focus areas and reducing burnout. The piece is largely commentary and implies a constructive long-term outlook for healthcare AI.

Analysis

The first-order implication is not that clinicians adopt more AI, but that utilization shifts from generic productivity tools to workflow-embedded decision support. That tends to benefit the layer closest to the clinician interface: EHR vendors, ambient documentation, triage, imaging, and clinical workflow software, while leaving stand-alone point solutions vulnerable unless they can prove measurable time savings or reimbursement lift. The second-order winner is likely software that reduces administrative load more than diagnostic tools, because the ROI is easier to quantify and procurement cycles are shorter.

The biggest economic effect is margin expansion via labor compression, but only over a multi-year horizon. Health systems are under structural staffing pressure, so even modest AI adoption that saves 5-10 minutes per encounter can translate into meaningful capacity gains without adding headcount. That creates a flywheel: less burnout improves retention, which lowers locum and overtime spend, which frees capital for further software spend. Vendors that can show hard metrics on throughput and documentation time should outperform generic “AI” narratives.

The contrarian view is that the market may be underestimating implementation friction rather than overestimating model quality. In healthcare, integration, liability, and workflow change-management are the bottlenecks; adoption can look exponential in surveys yet remain economically small for 12-24 months. The risk to the theme is a few high-profile clinical errors, reimbursement pushback, or hospital CIOs pausing spend until governance frameworks are standardized. In that sense, the near-term trade is less about AI hype and more about which incumbents can distribute and monetize AI safely inside regulated workflows.

Key Decisions for Investors

• Overweight EHR/workflow platforms with distribution leverage into clinicians (e.g., ORCL/EPIC-adjacent public proxies where available, or large-cap healthcare IT names) on a 6-12 month horizon; best risk/reward is for vendors that can convert AI features into higher ARPU rather than standalone startups.
• Long healthcare IT leaders vs short generic AI software basket: pair long healthcare workflow/enabling software against short names where the thesis depends on standalone clinician adoption; expect dispersion over 3-9 months as buyers demand proof of ROI.
• Buy 6-12 month call spreads on healthcare AI enablers after pullbacks, not into strength; the option structure fits a slow adoption curve with event-driven upside from hospital contract wins and product launches.
• Avoid chasing pure-play diagnostic AI until reimbursement and liability clarity improve; if forced to express bullishness, prefer picks-and-shovels exposure to data integration and documentation automation, where payback periods are easier to verify.
• Set a catalyst watchlist for health system earnings and IT spend commentary over the next 1-2 quarters; any evidence of budget reallocation toward AI-enabled productivity tools is the signal to add risk.