Labcorp announced on February 23, 2026 that it is expanding its long-running collaboration with computational pathology company PathAI to deploy PathAI’s AISight Dx digital pathology platform across Labcorp’s national network of anatomic pathology laboratories and affiliated hospital sites. The move takes a partnership that began in 2019 and turns it into a nationwide buildout, putting an FDA-cleared, AI-enabled diagnostic workflow in front of pathologists who read biopsy and tissue samples for cancer and other diseases every day. Financial terms were not disclosed.
What AISight Dx Actually Does
AISight Dx is PathAI’s cloud-native image management system, built to let pathologists view, annotate and collaborate on digitized slides instead of squinting through a microscope at physical glass. The platform received FDA clearance for primary diagnosis use in the United States in June 2025 and carries CE-IVD marking for primary diagnosis in the European Economic Area, the United Kingdom and Switzerland. Beyond slide viewing, it embeds AI-based image analysis tools directly into the workflow, meaning a pathologist reviewing a case can get algorithmic input on tumor detection, tissue quality or cellularity without leaving the platform. For Labcorp, the appeal is standardization at scale: the same digital infrastructure and AI tooling across every lab in the network, rather than a patchwork of local processes.
How Pathology Got Here
For more than a century, cancer diagnosis has depended on a pathologist examining thin slices of tissue on glass slides under a microscope, a process that is accurate but slow, dependent on the individual reader, and hard to scale as case volumes grow. The shift to “digital pathology” — scanning slides into high-resolution images that can be stored, shared and analyzed by software — only became commercially practical in the past decade. Paige, a Memorial Sloan Kettering spinout now owned by Tempus AI, became the first company to win FDA clearance for an AI-based primary diagnosis application in pathology in 2021, for prostate cancer detection. Since then, the FDA has cleared roughly two dozen AI-based cancer diagnostic tools, including PaigeProstate Detect, Stratipath Breast and Visiopharm Metastasis Detection, turning what was an experimental niche into an expanding category of cleared clinical software.
The Numbers Driving Adoption
The case for bringing AI into the pathology lab rests on studies showing it can match or beat unaided human readers on specific, narrow tasks. Research on a deep-learning model for diffuse large B-cell lymphoma found it achieved 100% diagnostic accuracy at two hospitals and 99.71% at a third, with 100% sensitivity, compared with 74.39% accuracy for human pathologists reading the same cases, according to findings reported in Frontiers in Oncology. In metastasis detection, a convolutional neural network framework reached 92.4% sensitivity, ahead of the best prior automated method at 82.7% and well ahead of human pathologists at 73.2%, per reporting from Labmedica. Those gaps, concentrated in high-volume, pattern-recognition-heavy tasks, are the specific niches where health systems are choosing to deploy AI first rather than asking it to replace a pathologist’s full judgment.
Beyond Labcorp: A Pattern Across Health Systems
The Labcorp deal is one of several PathAI deployments announced within a few months of each other. In January 2026, PathAI and University Hospital Zurich announced a collaboration to deploy AISight Dx along with the AIM-TumorCellularity algorithm for routine molecular pathology workflows, described as one of the first implementations of AI-based pathology tools in daily clinical operation in Switzerland. In April 2026, PathAI and MedStar Health announced a multi-year partnership to deploy AISight Dx along with algorithms including ArtifactDetect and TumorDetect across a network supporting more than 40 pathologists. Together with the Labcorp expansion, the three deals suggest 2026 has become the year digital pathology platforms moved from pilot programs at single academic centers to multi-site, production deployments inside major commercial and hospital lab networks.
The Skeptics: Liability, Jobs and a Digitization Gap
Not everyone in the field is convinced the technology is ready for the scale being proposed. Surveys of practicing pathologists have found that nearly half worry AI adoption could eventually displace jobs, even as most insist the technology should remain an assistive tool rather than a decision-maker. The liability question looms over every deployment: if an AI tool misses a diagnosis on a case a pathologist signs off on, legal and ethical responsibility still falls on the physician, not the software. Discussion at the 2026 USCAP pathology conference underscored a more basic constraint — only about 10% of pathology workflows in the United States are fully digitized today, and no single AI model handles the full variability of tissue types, staining methods and scanner outputs across labs, meaning scaling tools like AISight Dx nationally is a slower, harder engineering problem than the press releases imply.
What’s Next
“Labcorp is committed to building a modern, AI-powered infrastructure that sets a new standard for efficiency, collaboration and innovation in pathology,” said Dr. Marcia Eisenberg, Labcorp’s chief scientific officer, in the companies’ announcement. PathAI co-founder and CEO Dr. Andy Beck said the deployment “brings high-quality, efficient digital pathology to a national scale.” Whether that promise holds will depend on execution across hundreds of individual lab sites, each with its own scanners, case mix and pathologist staffing. With Leica Biosystems clearing new AI-assisted quality control software in August 2026 and Artera winning clearance for an AI breast cancer risk-stratification tool in May 2026, Labcorp’s rollout lands in the middle of a regulatory wave rather than ahead of it — suggesting that 2026’s real test for digital pathology AI is not whether the FDA will clear more tools, but whether labs can actually run them reliably at the scale Labcorp is now attempting.
Photo: ptra / PIXABAY via Pixabay
