Will AI Replace Pathologists' Assistants? AI Reads Slides — But Cannot Hold a Scalpel

AI can now analyze a histology slide and identify cancerous cells with accuracy that rivals — and sometimes exceeds — a trained pathologist. [Claim] That fact has generated enormous excitement and anxiety in the medical field, with headlines suggesting that pathology is on the verge of disruption. But if you are a pathologist's assistant, here is what those headlines miss: nobody is automating the part where you cut open the specimen, ink the margins, or decide which section of a tumor will go on the slide in the first place.

Pathologists' assistants face a 22% automation risk and 45% overall AI exposure in 2025. [Fact] Those numbers sit in a fascinating middle ground — high enough to matter, low enough to provide real job security — and the task-level breakdown explains exactly why a profession that supports one of the most AI-disrupted medical specialties remains remarkably stable itself.

The Physical-Digital Divide

Analyzing tissue specimens and documenting gross findings sits at 52% automation. [Fact] This is the most AI-exposed task in the role, and for good reason. AI-powered digital pathology tools can analyze scanned tissue images, flag abnormalities, measure tumor margins, and even suggest preliminary diagnoses. Companies like Paige.AI and PathAI have developed FDA-cleared algorithms that can detect prostate cancer, breast cancer, and other malignancies from digital slides with high sensitivity. [Claim] When the specimen is digital — a scanned slide, a photographed gross specimen — AI excels.

But performing gross dissection of surgical and autopsy specimens is at just 10% automation. [Fact] This is the hands-on core of the job: receiving a surgical specimen wrapped in saline-soaked gauze, orienting it anatomically using the surgeon's sutures as landmarks, inking the resection margins with different colors to track orientation through processing, dissecting it carefully to expose the pathology, and selecting the sections that will be processed for microscopy. Every specimen is different. Every tumor has a unique shape, position, and relationship to surrounding tissue. The pathologist's assistant must make real-time decisions about where to cut, what to sample, and how to preserve the diagnostic integrity of the tissue.

Consider a complex case: a partial colectomy specimen with a tumor near the surgical margin. The PA must measure tumor distance to the nearest inked margin to the nearest millimeter, identify and isolate regional lymph nodes from the surrounding fat (twelve or more is the standard for accurate staging), section the tumor through its thickest dimension to capture invasion depth, and document the entire process with photographs and detailed gross descriptions. Each of these decisions affects the cancer staging that determines the patient's treatment.

No robot is doing that. Not in 2025, and not by 2028 either. [Claim] The dexterity required to handle slippery, blood-soaked tissue, the spatial reasoning to orient a complex three-dimensional specimen, and the judgment to know which areas of a tumor are diagnostically critical all exceed current robotic capabilities by enormous margins.

Preparing and processing tissue sections for histological examination comes in at 35% automation. [Fact] Automated tissue processors and embedding stations handle some of the mechanical steps — overnight processing schedules, paraffin infiltration, microtome calibration. But quality control — ensuring proper fixation depth, correct orientation during embedding so the diagnostic surface faces the blade, and appropriate sectioning thickness — still requires trained human oversight. A poorly oriented block can render an entire diagnostic specimen useless and force a re-cut from the gross specimen, costing days of turnaround time.

Assisting at autopsies represents another largely manual domain at 15% automation. [Fact] External examination, organ removal, weight measurement, and the systematic documentation of pathology require physical presence and tactile assessment that no current technology approaches.

A Tiny but Growing Profession

With only approximately 2,800 pathologists' assistants in the U.S., this is one of the smallest occupations we track. [Fact] The BLS projects +7% growth through 2034, reflecting strong demand driven by an aging population generating more surgical pathology specimens and a national shortage of pathologists who need support staff. [Fact]

The median annual wage of $93,680 makes this one of the best-compensated allied health professions. [Fact] The specialized training required — typically a master's degree from one of the approximately 14 NAACLS-accredited pathologists' assistant programs in the U.S. — creates a barrier to entry that also protects against both automation and labor market competition. Students complete roughly 22 months of intensive training including surgical pathology rotations, autopsy experience, and forensic pathology exposure. That investment in human expertise is not easily replicated by software.

The professional landscape is also shaped by regulatory and accreditation requirements. The American Association of Pathologists' Assistants (AAPA) and the American Society for Clinical Pathology (ASCP) provide credentialing that hospitals and laboratories require. Board-certified PAs are increasingly seen as essential members of pathology departments rather than substitutes for pathologists — a positioning that reinforces job security against automation.

Why AI Actually Increases Demand for PAs

Here is the counterintuitive part: as AI makes digital pathology analysis faster and more accessible, pathology labs are processing more specimens, not fewer. [Claim] When AI can screen a slide in seconds, labs can accept higher volumes. Higher volumes mean more specimens need to be grossed, dissected, and prepared — the physical tasks that pathologists' assistants perform. The bottleneck in modern pathology is shifting from microscope time to specimen handling, and that bottleneck only resolves with more skilled human hands.

AI is also enabling pathologists to work remotely through digital slide review, which means the pathologist may not be physically present in the lab. That makes the on-site pathologist's assistant even more essential as the person who handles the physical specimen work, communicates with surgeons, ensures sample quality, and acts as the eyes and hands of remote pathologists. [Claim] Several large academic centers and reference laboratories have already restructured their workflows around this model, with PAs handling all gross work while pathologists at home or in distant cities perform microscopic review through digital pathology platforms.

There is also a secondary effect: as AI handles more of the screening work, pathologists are spending proportionally more time on complex cases that require sign-out judgment. This creates downstream demand for PAs to support more sophisticated grossing — including the careful documentation and sampling required for clinical trial specimens, molecular pathology workflows, and personalized medicine protocols.

The 2028 Outlook

By 2028, overall exposure is projected to reach 59% with automation risk at 34%. [Estimate] The increase will come almost entirely from improvements in digital specimen analysis and documentation tools — better voice-to-text gross dictation, AI-assisted measurement of tumor dimensions from photographs, and automated lymph node counting from scanned slides. The physical dissection and preparation tasks will remain largely unchanged because the technology to automate them simply does not exist in any practical form, and no current research program is aimed at building it.

What is likely to change is how PAs interact with technology. Voice-controlled gross dictation systems, augmented-reality measurement tools that overlay dimensions on photographed specimens, and AI-assisted sample selection algorithms that recommend tissue sections based on imaging findings are all in development or early deployment. Pathologists' assistants who master these tools will work faster, document more thoroughly, and provide more value to their pathology teams.

What This Means for Your Career

If you are a pathologist's assistant or considering this career, the data paints an encouraging picture: strong wage growth, positive employment outlook, and a physical skill set that AI complements rather than replaces. The combination of a master's-level educational barrier, hands-on physical work, and tight integration with the broader healthcare system creates a defensible professional position.

Three concrete recommendations stand out. First, invest in learning digital pathology tools and AI-assisted dictation systems — the PAs who handle these technologies effectively will become the most valuable members of their departments. Second, develop expertise in specialized grossing techniques for molecular pathology and clinical trials, where complex sample handling protocols are creating premium-skill niches. Third, consider leadership tracks: as pathology departments grow more technologically complex, experienced PAs are increasingly moving into supervisor and laboratory manager roles where they coordinate the human-AI workflow at the operational level.

That combination of physical expertise and digital literacy will define the top performers in this field. See the full analysis at [Pathologists' Assistants.]

---

AI-assisted analysis based on data from the Anthropic economic impact study, BLS occupational projections, and ONET task databases.\*