Pharmacy Technicians
29-2052.00Median wage $45,750/yr471,680 employed (US)Rank #85 of 923 scored · top 9% by substitution
Prepare medications under the direction of a pharmacist. May measure, mix, count out, label, and record amounts and dosages of medications according to prescription orders.
Sub-scores
0–100 · band = confidence interval from rater disagreement
Substitution49
Exposure50
Augmentation58
Substitution — the headline: capability discounted by cost, barriers and adoption.
Exposure — technical capability alone, regardless of whether anyone deploys it.
Augmentation — how much AI assists without replacing. High here + moderate substitution = a changing job, not a disappearing one.
Tasks on the substitution scale
21 rated tasks, binned by substitution score.
Position among all scored occupations
Distribution of 923 occupation scores; the marker is this occupation.
Tasks with substitution ≥ 70
33%
Run 1.0.0-draft.1 · computed 2026-08-05 · rater panel: claude-sonnet-5, claude-haiku-4-5-20251001 · intervals span rater disagreement.
Why this score
The five weighted dimensions of the composite, averaged across this occupation's tasks (importance-weighted, panel mean). Exact weights and formulas: /api/v1/methodology.
Task automatabilityw 35%50
panel mean rating 3.0/5 → substitution pressure 50/100
Technical feasibility todayw 20%51
panel mean rating 3.0/5 → substitution pressure 51/100
Cost vs. human wagew 15%54
panel mean rating 3.2/5 → substitution pressure 54/100
Adoption barriersw 20%inverted — strong barriers lower the score43
panel mean rating 3.3/5 (barrier strength) → substitution pressure 43/100
Sector adoption velocityw 10%48
panel mean rating 2.9/5 → substitution pressure 48/100
Task breakdown (21 tasks)
Substitution pressure per task, weighted by O*NET importance in the composite. Expand a task for the full rater audit trail — every rating, every model, every rationale.
Compute charges for medication or equipment dispensed to hospital patients and enter data in computer.
84CI 75–92 · exposure 87 · augmentation 75 · importance 4.7/5 · click for rater detail
Compute charges for medication or equipment dispensed to hospital patients and enter data in computer.
84| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 4/5 | Hospital pharmacy departments are highly digitized and actively adopting automation for routine back-office tasks like billing and data entry. Major health systems have already implemented automated charge capture and billing systems, reflecting rapid adoption in the healthcare/information sector. |
| Sector adoption velocity | claude-sonnet-5 | 4/5 | Healthcare administrative/billing functions have seen substantial automation adoption via EHR and pharmacy management systems, though full AI-driven automation lags behind finance/tech sectors. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI systems assist pharmacy technicians by automatically calculating charges and pre-populating billing records, allowing technicians to focus on verification, exception handling, and complex cases. This augmentation significantly raises technician productivity while keeping human oversight in place. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI-enabled software significantly speeds up charge computation and reduces data entry errors, letting technicians focus on verification and exception handling. |
| Task automatability | claude-haiku-4-5-20251001 | 5/5 | This task is highly structured and repetitive: it involves applying predetermined pricing rules to dispensed items and entering standardized data into a computer system. Current AI systems can extract medication/equipment details from pharmacy records, apply billing logic, and populate EHR or billing systems with minimal human intervention, easily meeting the 50% time-saving threshold. |
| Task automatability | claude-sonnet-5 | 4/5 | Computing charges and entering data into pharmacy/billing systems is a structured, rules-based task well suited to automation via pharmacy management software and RPA/AI integration, though system integration and exception handling still require some setup. |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | While charge accuracy has compliance implications and some oversight is advisable, there are no hard legal requirements that a licensed human must sign off on every charge entry. Hospitals routinely automate this process with standard audit controls, creating only modest friction for full substitution. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensure is required specifically for billing/data entry, but hospital systems have institutional and workflow inertia, plus some oversight expectations tied to accuracy of medication records. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | The cost of inference and integration for pharmacy billing automation is negligible per transaction compared to the loaded hourly wage of a pharmacy technician ($30–40/hour fully loaded), making automated charge calculation and entry orders of magnitude cheaper. |
| Cost vs. human wage | claude-sonnet-5 | 4/5 | Automated charge computation and data entry via existing software is far cheaper per transaction than manual technician time, though licensing and integration costs keep it from being a full order-of-magnitude cheaper in all settings. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 5/5 | Pharmacy management and billing software with integrated automation (barcode scanning, auto-pricing, claims processing) is deployed at scale across hospital systems today. Multiple mature products (e.g., integrated pharmacy information systems) reliably perform charge computation and data entry in production environments. |
| Technical feasibility today | claude-sonnet-5 | 4/5 | Deployed pharmacy information systems and hospital billing software already automate charge capture and data entry in production at many hospitals, though edge cases (insurance overrides, formulary exceptions) still require human review. |
Price stock and mark items for sale.
81CI 79–84 · exposure 75 · augmentation 50 · importance 3.6/5 · click for rater detail
Price stock and mark items for sale.
81| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 4/5 | Large pharmacy chains and retail organizations have already deployed automated pricing and label-printing systems in production at scale; adoption is rapid in high-volume, digitized retail environments typical of pharmacy operations. |
| Sector adoption velocity | claude-sonnet-5 | 4/5 | Retail pharmacy chains have widely adopted automated inventory and pricing systems as part of standard POS and supply chain software. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI/automation assists technicians by generating accurate labels and flagging pricing discrepancies, reducing manual lookup and writing work, though human review of exceptions and overrides remains normal workflow practice. |
| Augmentation potential | claude-sonnet-5 | 3/5 | Where not fully automated, software assists technicians by suggesting prices and flagging discrepancies, moderately boosting efficiency. |
| Task automatability | claude-haiku-4-5-20251001 | 4/5 | Pricing and marking items for sale is highly automatable: barcode scanning, inventory database lookup, and price label generation can be fully automated with barcode printers. The core cognitive work (looking up price, applying markup rules) is algorithmic and repeatable, easily achieving >50% time savings at equal quality. |
| Task automatability | claude-sonnet-5 | 4/5 | Pricing and marking items for sale is a rules-based, repetitive data task that pharmacy inventory/POS systems can already automate via barcode scanning and integrated pricing databases with minimal human intervention. |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | Few barriers exist: pricing is not a licensed task, though pharmacy regulation and chain policies may mandate human oversight of certain price changes. Most friction is organizational preference for quality control rather than legal requirement. |
| Adoption barriers | claude-sonnet-5 | 1/5 | No licensing or regulatory requirement mandates a human perform pricing/stocking; it's a purely administrative/inventory function. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | The cost of barcode scanning and label printing is negligible per item compared to pharmacist or technician labor; a fully automated system (one-time setup plus marginal inference cost) is orders of magnitude cheaper than manual pricing and marking. |
| Cost vs. human wage | claude-sonnet-5 | 5/5 | Automated pricing/labeling software and barcode systems cost far less per transaction than manual technician labor for this repetitive task. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 4/5 | Mature retail and pharmacy systems (e.g., pharmacy management software integrated with label printers) reliably perform inventory pricing and marking at scale in production environments today. Error rates are low for standard items, though edge cases (manual overrides, special pricing) still require human review. |
| Technical feasibility today | claude-sonnet-5 | 4/5 | Retail and pharmacy management systems (e.g., automated inventory/pricing modules integrated with POS) already perform this reliably in production across chain pharmacies today. |
Operate cash registers to accept payment from customers.
77CI 61–92 · exposure 80 · augmentation 63 · importance 4.5/5 · click for rater detail
Operate cash registers to accept payment from customers.
77| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 4/5 | Major pharmacy chains and retailers have aggressively deployed self-checkout and automated payment systems over the past decade, with measurable technician hour displacement in high-volume locations. |
| Sector adoption velocity | claude-sonnet-5 | 4/5 | Retail and pharmacy sectors have broadly adopted self-checkout and automated payment systems, though full replacement of staff varies by store size and format. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI-powered POS systems assist technicians by automatically applying discounts, processing insurance claims in real-time, and flagging drug-interaction alerts tied to payment, significantly raising efficiency while the technician remains present for verification and customer service. |
| Augmentation potential | claude-sonnet-5 | 3/5 | POS systems assist technicians with faster, more accurate transaction processing, though the task itself is simple and augmentation value is moderate. |
| Task automatability | claude-haiku-4-5-20251001 | 3/5 | Modern point-of-sale systems and self-checkout kiosks powered by AI can handle payment processing, inventory deduction, and cash/card validation, though integration with existing pharmacy systems and handling of edge cases (return authorizations, insurance coordination) requires human oversight, achieving ~50% time savings in routine transactions. |
| Task automatability | claude-sonnet-5 | 5/5 | Payment acceptance via cash registers/POS is already largely automated through self-checkout, card terminals, and automated kiosks widely deployed in pharmacies and retail. |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | Customers often prefer human interaction for medication questions and trust; regulatory requirements around controlled substances and data privacy add friction, though no hard legal mandate requires a human to operate the register specifically. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement for handling payments; some friction exists from customer preference for human interaction and loss-prevention concerns, but no legal barrier. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 4/5 | AI-powered self-checkout systems cost significantly less per transaction than a full-time technician wage ($16–22/hour) when amortized across high transaction volumes, though initial capital and maintenance costs must be factored in. |
| Cost vs. human wage | claude-sonnet-5 | 5/5 | Automated payment terminals and self-checkout kiosks cost far less per transaction than a human cashier's wage, especially at scale. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 4/5 | Self-checkout and AI-assisted POS systems are widely deployed in retail and pharmacy settings (CVS, Walgreens, Walmart) and operate reliably for straightforward transactions, though some error rates and the need for human intervention on exceptions prevent a perfect 5 rating. |
| Technical feasibility today | claude-sonnet-5 | 5/5 | Self-checkout and automated POS systems are mature, deployed at scale in pharmacies and retail chains today, reliably processing payments. |
Price and file prescriptions that have been filled.
72CI 70–75 · exposure 75 · augmentation 50 · importance 4.4/5 · click for rater detail
Price and file prescriptions that have been filled.
72| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 4/5 | Large pharmacy chains (CVS, Walgreens, Rite Aid) and hospital systems have widely deployed automated pricing and filing systems over the past 5+ years. Smaller independent pharmacies lag, but overall penetration in the digitized pharmacy sector is substantial, with continued expansion in primary care and mail-order environments. |
| Sector adoption velocity | claude-sonnet-5 | 4/5 | Retail and hospital pharmacy chains have already deeply integrated automated pricing/insurance adjudication and electronic filing into standard workflows, representing mature, widespread adoption. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI assists pharmacy technicians by automating routine lookups and filing, freeing them to focus on verification, customer service, and exception handling. While not transformative, it meaningfully raises output per technician on this narrow, repetitive task. |
| Augmentation potential | claude-sonnet-5 | 3/5 | Where not fully automated, software still assists technicians by pre-populating pricing and filing data, reducing manual lookup and entry effort. |
| Task automatability | claude-haiku-4-5-20251001 | 4/5 | Current AI and RPA systems can reliably extract prescription data, match it to pricing databases, and file records in pharmacy management systems with high accuracy. The task involves rule-based data entry and workflow routing with minimal judgment, allowing >50% time savings with off-the-shelf pharmacy software integrations. |
| Task automatability | claude-sonnet-5 | 4/5 | Pricing via insurance adjudication systems and electronic filing/recordkeeping are highly structured, rules-based tasks already handled by pharmacy management software with minimal human intervention beyond exception handling. |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | Moderate barriers exist: HIPAA compliance and audit trails require careful system design, insurance plan integrations vary, and some state regulations require pharmacist oversight of certain filings. Customer preference for human contact and organizational inertia in smaller pharmacies also slow adoption, but no absolute legal prohibition on automation exists. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensure is strictly required for this specific pricing/filing sub-task, but pharmacy operations are embedded in regulated recordkeeping systems requiring accuracy and audit compliance, creating some institutional friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 4/5 | Automated prescription pricing and filing costs roughly $0.10–0.50 per transaction (software licenses and integration), compared to a pharmacy technician burdened cost of $15–25 per hour for the 5–10 minutes this task typically requires, yielding 10:1 to 30:1 cost advantage for automation. |
| Cost vs. human wage | claude-sonnet-5 | 4/5 | Automated adjudication and digital filing systems process transactions at a fraction of the marginal labor cost per prescription, though software licensing and integration costs are nontrivial. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 4/5 | Deployed pharmacy management systems and RPA vendors (including pharmacy-specific solutions from major EHR vendors) routinely automate pricing lookup and electronic filing in production environments. Error rates are low for standard prescriptions, though edge cases and insurance denials still require human review. |
| Technical feasibility today | claude-sonnet-5 | 4/5 | Pharmacy management systems (e.g., automated claims adjudication, electronic prescription records) already perform pricing and filing reliably in production across most retail and hospital pharmacies today. |
Establish or maintain patient profiles, including lists of medications taken by individual patients.
72CI 70–74 · exposure 75 · augmentation 75 · importance 4.7/5 · click for rater detail
Establish or maintain patient profiles, including lists of medications taken by individual patients.
72| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 4/5 | Large pharmacy chains and hospital systems are rapidly deploying EHR-integrated automation for medication reconciliation and profile maintenance. Smaller independent pharmacies lag, but the trend is strong in chains and health systems—the dominant employment segment. |
| Sector adoption velocity | claude-sonnet-5 | 4/5 | Retail and hospital pharmacy chains have already deeply integrated automated patient profile management and e-prescribing systems, reflecting fast, sector-wide adoption of digitized records. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI-assisted profile management dramatically reduces technician time on data entry, allowing them to focus on patient interaction, verification, and exception handling. The human stays in the loop for clinical judgment while AI handles routine capture and maintenance. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI-assisted systems flag interactions, auto-suggest updates, and reduce manual lookup burden, meaningfully boosting technician efficiency while they retain responsibility for verification. |
| Task automatability | claude-haiku-4-5-20251001 | 4/5 | This task is highly structured data entry and retrieval from electronic health records (EHRs). Current AI systems can extract medication information from prescriptions, insurance claims, and clinical notes with high accuracy, automatically populate patient profiles, and flag inconsistencies—delivering >50% time savings at equal quality in most cases. Manual verification by pharmacy staff remains necessary, but the bulk work is automatable. |
| Task automatability | claude-sonnet-5 | 4/5 | Compiling and updating medication lists from structured pharmacy records is a data-entry/data-management task that current AI and pharmacy software can largely automate, though edge cases (reconciling conflicting records, allergy flags) need human review. |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | Pharmacies must comply with HIPAA and state pharmacy regulations, and final verification by a licensed pharmacist is often required, adding oversight friction. However, no law explicitly forbids AI from creating or updating profiles; the barrier is organizational/regulatory rigor rather than legal prohibition. |
| Adoption barriers | claude-sonnet-5 | 3/5 | Patient health data handling triggers HIPAA/privacy and pharmacy recordkeeping regulations, and errors in medication lists carry safety liability, creating moderate compliance and oversight barriers even though the task itself isn't reserved to licensed judgment. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | EHR automation and AI-driven medication extraction cost mere cents per patient profile per month, while a pharmacy technician's loaded labor cost is $25–35/hour. Automated systems are an order of magnitude cheaper when amortized across patient populations. |
| Cost vs. human wage | claude-sonnet-5 | 4/5 | Automated database/EHR integration and AI-assisted data entry cost far less per record than technician time spent manually establishing and updating profiles, though system licensing and oversight add some cost. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 4/5 | Mature EHR systems and pharmacy management software (e.g., Accredo, Omnicell, pharmacopoeia databases integrated with AI) already perform significant portions of this task in production pharmacies. OCR and NLP models reliably extract medication lists from images and documents; profile maintenance is increasingly automated, though integration varies by pharmacy chain. |
| Technical feasibility today | claude-sonnet-5 | 4/5 | Pharmacy management systems already auto-populate and update patient medication profiles from prescription and insurance data feeds, with pharmacy technicians verifying rather than manually building most records. |
Enter prescription information into computer databases.
71CI 67–74 · exposure 75 · augmentation 88 · importance 4.8/5 · click for rater detail
Enter prescription information into computer databases.
71| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 3/5 | Pharmacy chains and hospital systems have piloted automated prescription entry, but adoption remains uneven. Many smaller independent pharmacies and rural practices still rely on manual entry, and regulatory caution has slowed full end-to-end automation without human sign-off, placing this in the "pilot and early deployment" range rather than mature saturation. |
| Sector adoption velocity | claude-sonnet-5 | 4/5 | Healthcare and retail pharmacy have widely adopted e-prescribing and automated data entry systems, driven by regulatory push (e.g., EPCS) and efficiency demands, though full end-to-end automation still varies by pharmacy size. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI-assisted form-filling and auto-population of common fields meaningfully accelerate data entry while pharmacists and technicians retain oversight and error-checking roles. This creates clear productivity gains without removing human responsibility for accuracy. |
| Augmentation potential | claude-sonnet-5 | 5/5 | AI-assisted data entry, autofill, and error-flagging tools significantly speed up technician workflows while keeping humans in the loop for verification and exception handling. |
| Task automatability | claude-haiku-4-5-20251001 | 4/5 | Prescription data entry is highly structured and rule-based, with clear fields (patient ID, drug name, dosage, quantity, refills). Current OCR + NLP systems can extract this from handwritten or printed prescriptions and populate databases reliably, achieving substantial time savings, though some edge cases (unusual abbreviations, illegible writing) may require human review. |
| Task automatability | claude-sonnet-5 | 4/5 | Data entry from structured or semi-structured prescription sources (e-prescriptions, scanned forms) is highly automatable via OCR, NLP extraction, and integration with pharmacy management systems, though verification of edge cases still needs human review. |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | Pharmacy regulations require accurate medication records and often mandate human verification of prescriptions before dispensing, creating operational friction. However, automation of data entry itself (as opposed to verification) is not legally blocked, so barriers are moderate rather than absolute. |
| Adoption barriers | claude-sonnet-5 | 3/5 | Pharmacy regulations often require technician or pharmacist verification of entered data for safety and legal compliance, creating oversight requirements even if the initial entry is automated. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | AI inference for OCR and database entry costs cents per prescription, while pharmacy technician labor (burdened wage ~$20–30/hour) costs several dollars per entry. The cost advantage is one to two orders of magnitude in favor of automation. |
| Cost vs. human wage | claude-sonnet-5 | 4/5 | Automated data capture and integration software costs a fraction of technician labor hours per entry once implemented, though initial system setup and periodic human verification add some cost. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 4/5 | Multiple deployed pharmacy systems and general data-entry automation tools demonstrate this capability in production environments. Pharmacy chains and healthcare systems already use automated prescription entry with OCR and form-filling, though they typically retain human verification as a safeguard rather than relying fully on automation. |
| Technical feasibility today | claude-sonnet-5 | 4/5 | E-prescribing systems and pharmacy software already auto-populate most prescription data directly from prescriber systems, and OCR/NLP tools handle faxed or handwritten scripts in production, though accuracy issues remain for unusual formats. |
Receive and store incoming supplies, verify quantities against invoices, check for outdated medications in current inventory, and inform supervisors of stock needs and shortages.
71CI 55–86 · exposure 67 · augmentation 75 · importance 4.6/5 · click for rater detail
Receive and store incoming supplies, verify quantities against invoices, check for outdated medications in current inventory, and inform supervisors of stock needs and shortages.
71| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 5/5 | Pharmacy is a highly digitized, information-intensive sector with strong economic pressure to reduce labor costs. Inventory automation systems are already widely deployed across hospital and retail pharmacy chains, with rapid ongoing adoption. |
| Sector adoption velocity | claude-sonnet-5 | 3/5 | Healthcare/retail pharmacy sectors have moderate digitization with barcode and inventory systems common in larger chains, but adoption is uneven across independent and hospital pharmacies. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI systems augment technician productivity significantly by automating routine counting and checking, freeing technicians for higher-value tasks like patient counseling and medication compounding. Real-time alerts and dashboards assist supervisory decision-making on stock management. |
| Augmentation potential | claude-sonnet-5 | 4/5 | Inventory management systems significantly help technicians track stock levels, flag expirations, and generate reorder reports, meaningfully boosting efficiency while humans still verify and act on the data. |
| Task automatability | claude-haiku-4-5-20251001 | 4/5 | Most components of this task are highly automatable: barcode scanning and invoice matching are fully automated in modern pharmacy management systems, inventory tracking via RFID/barcoding eliminates manual counts, and expiration date checking is automated through database queries. However, some physical handling and final supervisory judgment may still require human involvement, preventing a perfect 5. |
| Task automatability | claude-sonnet-5 | 3/5 | Inventory reconciliation, expiration checks, and reorder alerts can be substantially automated via barcode/RFID scanning and inventory management software, but physical receiving/unpacking and some judgment calls remain manual. |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | While pharmacy practice is regulated, inventory management itself is not a licensed professional task requiring pharmacist sign-off on routine stock verification and expiration checks. Integration with existing pharmacy systems and staff workflows creates mild adoption friction, but no legal barrier prevents automation. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement mandates a human specifically for stock-checking tasks, though pharmacy technicians are often regulated for medication-handling functions broadly, creating some indirect friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | The cost of AI-enabled inventory management software (per-transaction or per-facility licensing) is orders of magnitude lower than paying pharmacy technicians for routine stock verification and monitoring tasks, especially at scale across multiple locations. |
| Cost vs. human wage | claude-sonnet-5 | 3/5 | Inventory software and scanners have upfront and subscription costs but reduce labor hours; savings are real but not order-of-magnitude given hardware, integration, and technician oversight still needed. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 5/5 | Pharmacy management systems (e.g., Omnicell, BD Pyxus, QS/1) deployed in thousands of pharmacies worldwide already perform inventory tracking, automated receipt verification, and expiration date monitoring. These are mature, production-scale systems used daily in healthcare settings. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | Pharmacy inventory management systems with automated expiration tracking and reorder alerts are deployed in many pharmacies, but physical handling and verification still require human involvement, and smaller pharmacies often lack full automation. |
Order, label, and count stock of medications, chemicals, or supplies and enter inventory data into computer.
69CI 64–75 · exposure 75 · augmentation 63 · importance 4.4/5 · click for rater detail
Order, label, and count stock of medications, chemicals, or supplies and enter inventory data into computer.
69| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 4/5 | Large hospital systems and retail pharmacy chains have already invested heavily in robotic automation for dispensing and inventory; adoption is production-grade in institutional settings. Smaller independent pharmacies lag, but the overall trajectory in high-volume, digitized pharmacy operations shows deep and sustained automation deployment. |
| Sector adoption velocity | claude-sonnet-5 | 3/5 | Large retail and hospital pharmacy chains have adopted automated inventory and dispensing systems, but many independent pharmacies still rely on manual or semi-manual processes, giving a middling overall adoption picture. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | Barcode scanners and automated inventory tracking assist technicians by reducing manual counting errors and streamlining data entry, raising accuracy and speed. However, the augmentation is modest compared to full automation potential; humans remain primarily to oversee and verify system outputs rather than to perform cognitive work enhanced by AI. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI-enabled inventory software significantly boosts efficiency by predicting reorder needs, flagging discrepancies, and reducing manual counting errors, while technicians remain involved in verification and physical handling. |
| Task automatability | claude-haiku-4-5-20251001 | 4/5 | Most of this task—counting stock, applying labels, and entering inventory data—can be automated with barcode scanners, automated counting systems, and robotic pharmacy automation platforms. Ordering can be triggered automatically based on inventory thresholds. The main limitation is physical handling variability and occasional human judgment on stock rotation, but 50%+ time savings is readily achievable with deployed systems. |
| Task automatability | claude-sonnet-5 | 4/5 | Inventory ordering, labeling, and counting are highly structured, repetitive data-management tasks well-suited to automation via barcode scanning, RFID, and pharmacy inventory management systems integrated with automated reordering algorithms.4This can achieve substantial time savings, though physical counting/handling of stock still requires some human or robotic intervention. |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | Inventory management and restocking are not strictly regulated as to who performs them, but pharmacy operations are highly regulated environments where quality and accuracy standards create organizational friction. Most pharmacies maintain human oversight of automated systems and verify critical counts, creating moderate barriers to full substitution. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensure specifically requires a human to perform routine inventory counting/ordering, though pharmacy regulations around controlled substance tracking and recordkeeping create some compliance-related friction requiring verification steps. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 4/5 | Pharmacy automation systems have high upfront capital cost but very low per-unit operating cost once installed. For high-volume operations, the inference and integration cost per task becomes significantly cheaper than a loaded pharmacy technician wage; for small pharmacies the ratio is less favorable, but most deployment occurs in high-volume settings. |
| Cost vs. human wage | claude-sonnet-5 | 3/5 | While software licensing and automated dispensing hardware have upfront and maintenance costs, at scale they can be cheaper than technician labor for repetitive counting/ordering, though initial capital costs keep this from being an order-of-magnitude savings for smaller pharmacies. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 4/5 | Robotic pharmacy dispensing and inventory management systems are in production use at major hospitals and retail chains (e.g., Parata, ScriptPro, RxSafe). These systems reliably count, label, and track medications at scale. Some edge cases and manual oversight remain, but the core task is demonstrably performed reliably by deployed products. |
| Technical feasibility today | claude-sonnet-5 | 4/5 | Automated pharmacy inventory systems (e.g., automated dispensing cabinets, robotic dispensing systems, pharmacy management software with auto-reorder triggers) are already deployed widely in retail and hospital pharmacies today. |
Prepare and process medical insurance claim forms and records.
66CI 57–75 · exposure 70 · augmentation 75 · importance 4.7/5 · click for rater detail
Prepare and process medical insurance claim forms and records.
66| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 3/5 | Large pharmacy chains and PBMs have piloted/deployed claim automation, but adoption remains uneven; many smaller pharmacies and independents still rely on manual processing or older systems. Production deployment is growing but not yet industry-standard, placing this in middling-adoption territory. |
| Sector adoption velocity | claude-sonnet-5 | 4/5 | Retail and institutional pharmacy has widely adopted automated claims/PBM processing systems for years, representing deep, mature adoption in this specific workflow. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI can draft forms, flag missing fields, auto-populate from patient records, and highlight exceptions needing technician judgment—substantially raising technician productivity on triage and exception handling while keeping the human in the loop for final review and appeals. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI-assisted claims software significantly speeds up technicians' processing of routine claims and flags errors, while humans still manage denials and complex cases. |
| Task automatability | claude-haiku-4-5-20251001 | 4/5 | Claim form preparation and processing involves substantial repetitive data entry, form field mapping, and rule-based eligibility checking—tasks well-suited to current AI. OCR, structured data extraction, and rule engines can handle 70–80% of the workflow end-to-end, with remaining 20–30% requiring human review for exceptions, achieving >50% time savings. |
| Task automatability | claude-sonnet-5 | 4/5 | Insurance claim preparation is largely structured data entry and rule-based adjudication that current AI/automation systems handle well, though edge cases (rejected claims, prior authorizations) still need human review. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Insurance claim handling is heavily regulated under HIPAA, state insurance law, and payer contracts; there are strict audit and compliance requirements, and liability falls to the pharmacy/payer if claims are denied or delayed due to AI errors. These regulatory and contractual constraints significantly slow substitution. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensure requirement specifically for claims paperwork, but pharmacy operations still require technician/pharmacist oversight for exceptions, errors, and controlled substance-related claims. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 4/5 | OCR, extraction, and rule-based processing are cheap per transaction; integration and oversight overhead is moderate. Cost per claim processed is typically 1/5 to 1/10 of human technician cost for routine claims, though human review of exceptions adds back some cost. |
| Cost vs. human wage | claude-sonnet-5 | 4/5 | Automated claims processing software costs a small fraction of technician labor time per claim once integrated with pharmacy systems, though initial setup and edge-case handling add cost. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 3/5 | RPA and claim automation products exist and handle high-volume straightforward claims in production (e.g., pharmacy benefit manager systems), but error rates on complex cases (formulary exceptions, prior auth disputes) remain material; widespread reliable full-task automation is not yet proven at scale across all claim types. |
| Technical feasibility today | claude-sonnet-5 | 4/5 | Pharmacy management systems already have automated claims adjudication and e-prescribing integration in production at scale, though exceptions and appeals still route to technicians. |
Receive written prescription or refill requests and verify that information is complete and accurate.
59CI 45–74 · exposure 62 · augmentation 88 · importance 4.9/5 · click for rater detail
Receive written prescription or refill requests and verify that information is complete and accurate.
59| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 4/5 | Large pharmacy chains (CVS, Walgreens, Walmart) and PBMs have already deployed automated prescription verification and routing systems in high-volume settings. Smaller independent pharmacies lag, but major systems show rapid, production-level adoption in the digitized, high-volume pharmacy sector. |
| Sector adoption velocity | claude-sonnet-5 | 3/5 | Healthcare/pharmacy retail has moderate digitization with e-prescribing and automated verification tools deployed at scale in chains, but full agentic automation of this specific compliance-sensitive task remains limited. |
| Augmentation potential | claude-haiku-4-5-20251001 | 5/5 | AI dramatically augments technician productivity by automating routine data entry validation, flagging missing fields, detecting common errors, and routing prescriptions intelligently. Technicians remain in the loop for judgment calls, reducing cognitive load and error rates while allowing them to focus on complex or ambiguous cases. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI-assisted prescription parsing, drug interaction checks, and format validation meaningfully speed up technicians' review process while they remain responsible for final verification. |
| Task automatability | claude-haiku-4-5-20251001 | 4/5 | AI systems can reliably extract, parse, and validate prescription data from digital and scanned images, checking for completeness (patient ID, drug name, dosage, physician signature) and common errors. Current OCR and rule-based validation achieve high accuracy on structured prescription forms, though edge cases and handwritten prescriptions introduce friction that prevents a full 5 rating. |
| Task automatability | claude-sonnet-5 | 3/5 | OCR/NLP systems can extract and check prescription data fields, but validating completeness and accuracy against pharmacy law and clinical context still typically requires human verification, limiting full automation. |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | While many jurisdictions allow AI-assisted verification and triage, pharmacy licensing rules and liability concerns create moderate friction. Most U.S. states still require a licensed technician or pharmacist to sign off on prescription accuracy, and insurance/legal liability for errors incentivizes human oversight even where automation is permitted. |
| Adoption barriers | claude-sonnet-5 | 4/5 | Prescription verification is subject to state pharmacy regulations often requiring a licensed technician or pharmacist to confirm accuracy before dispensing, creating substantial regulatory and liability barriers to full automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | AI-powered prescription verification costs a fraction of a technician's hourly wage (typically $15–18/hour loaded). Once integrated into a pharmacy system, the per-prescription inference and validation cost is measured in cents, representing at least a 10–100× cost advantage. |
| Cost vs. human wage | claude-sonnet-5 | 3/5 | Automated data-entry and validation checks are cheap to run, but integration with pharmacy systems, exception handling, and required human oversight keep overall cost comparable to technician labor rather than dramatically cheaper. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 4/5 | Deployed prescription processing software and pharmacy management systems routinely perform data extraction and basic validation checks in production environments. Mature systems handle digital prescriptions and scanned documents with minimal manual review, though some ambiguous or handwritten prescriptions still require human verification. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | Pharmacy workflow software and e-prescribing systems already flag missing fields or format errors, but these are decision-support tools rather than fully autonomous verifiers, and technicians still review requests. |
Supply and monitor robotic machines that dispense medicine into containers and label the containers.
56CI 41–70 · exposure 67 · augmentation 75 · importance 4.4/5 · click for rater detail
Supply and monitor robotic machines that dispense medicine into containers and label the containers.
56| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 4/5 | Large-scale adoption is well underway in hospital and institutional pharmacies with high transaction volumes; smaller retail pharmacies adopt more slowly due to capital constraints, but the trend across digitized healthcare settings is clearly accelerating. |
| Sector adoption velocity | claude-sonnet-5 | 3/5 | Pharmacy automation adoption is moderate and growing in large retail and hospital pharmacies, but many smaller pharmacies still rely on manual processes, reflecting middling sector-wide adoption. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI-powered monitoring (predictive alerts, error detection, performance dashboards) significantly enhances technician productivity by reducing manual inspection time and enabling faster exception handling and system optimization. |
| Augmentation potential | claude-sonnet-5 | 4/5 | Robotic dispensing significantly boosts technician productivity by handling repetitive counting and labeling tasks, letting technicians focus on oversight, restocking, and exception management. |
| Task automatability | claude-haiku-4-5-20251001 | 4/5 | Robotic pharmacy systems (e.g., Omnicell, ScriptPro, Baxter) already automate much of the dispensing and labeling workflow. However, monitoring and troubleshooting failures, handling exceptions, and quality control still require human intervention, preventing a full 5. |
| Task automatability | claude-sonnet-5 | 3/5 | Robotic dispensing systems already handle much of the mechanical counting/filling work, but supplying stock, monitoring for errors, and troubleshooting still require human presence and judgment, limiting full end-to-end automation. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Pharmacy operations are heavily regulated (FDA, state boards); liability for dispensing errors creates strong incentive to keep human oversight, and pharmacists must legally certify accuracy, creating a hard requirement for human sign-off. |
| Adoption barriers | claude-sonnet-5 | 4/5 | Pharmacy dispensing is heavily regulated, requiring licensed personnel to oversee and verify dispensing accuracy, with legal liability for errors, creating strong barriers to full automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 4/5 | The capital and operating costs of robotic systems are high, but labor savings from reduced manual dispensing and labeling are substantial; per-dose costs are significantly lower than human equivalent once amortized. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | Robotic dispensing systems have high upfront capital and maintenance costs, and still require a technician to supply and monitor them, so per-task cost savings versus a human-only process are moderate rather than order-of-magnitude cheaper. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 5/5 | Robotic dispensing and labeling systems are deployed at scale in hospitals and large pharmacies today, with proven production reliability and integration into existing pharmacy workflows. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | Automated pharmacy dispensing robots (e.g., ScriptPro, Parata) are deployed in many pharmacies today, but they still require technician oversight, restocking, and exception handling, so it's not a fully autonomous product. |
Answer telephones, responding to questions or requests.
55CI 54–56 · exposure 50 · augmentation 63 · importance 4.6/5 · click for rater detail
Answer telephones, responding to questions or requests.
55| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 4/5 | Pharmacy chains (CVS, Walgreens, regional chains) are actively deploying IVR and call-handling automation as part of digital-first strategies. Adoption is measurably happening in production in large chains, though smaller independent pharmacies lag. |
| Sector adoption velocity | claude-sonnet-5 | 3/5 | Retail pharmacy and healthcare call centers are adopting AI voice agents at a moderate pace, with pilots and partial deployments more common than full-scale replacement. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI systems can assist technicians by pre-screening calls, retrieving patient and medication information, and suggesting responses, but the task itself is largely handled directly by either the human or the machine. Moderate augmentation value on efficiency and accuracy of information retrieval. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI can pre-triage calls, provide scripted answers, and route or summarize requests, meaningfully boosting technician efficiency while humans handle complex or sensitive queries. |
| Task automatability | claude-haiku-4-5-20251001 | 3/5 | AI can handle straightforward, scripted telephone inquiries about common questions (hours, refill status, basic information), but the task requires judgment for complex medication questions, insurance issues, or upset customers that still demand human intervention. Partial automation with 50% time savings is feasible for routine calls with proper routing. |
| Task automatability | claude-sonnet-5 | 3/5 | AI phone/chat agents can handle routine calls (refill status, hours, prescription pickup) but pharmacy calls often involve nuanced medical, insurance, or urgent clinical questions requiring escalation, limiting full end-to-end automation. |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | Pharmacies can legally deploy automated systems for routine inquiries, but many customers and pharmacists prefer human contact for medication questions, and liability concerns over misheard or misunderstood medical information create organizational friction. No licensing barrier exists for AI handling calls. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No strict licensing requirement to answer phones, but pharmacy communications involve HIPAA compliance, medication safety, and liability concerns that push some interactions back to trained staff. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 4/5 | Telephone automation (IVR, AI agents) scales at very low marginal cost per call once deployed, making it substantially cheaper than staffing a technician to answer calls full-time. However, integration and oversight overhead means not quite an order-of-magnitude difference. |
| Cost vs. human wage | claude-sonnet-5 | 4/5 | Automated IVR/voice-AI systems are much cheaper per call than a technician's time for routine inquiries like refill status or store hours. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 3/5 | IVR systems and AI chatbots deployed in some pharmacy chains today handle call routing and simple queries, but they still have notable error rates on medication-specific questions and require human escalation for non-standard requests. Production systems exist but have material limitations in scope and reliability. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | Conversational AI phone systems are deployed in retail and healthcare settings for routine queries, but pharmacy-specific deployments still have significant error rates and typically route complex requests to humans. |
Prepack bulk medicines, fill bottles with prescribed medications, and type and affix labels.
54CI 41–67 · exposure 62 · augmentation 75 · importance 4.3/5 · click for rater detail
Prepack bulk medicines, fill bottles with prescribed medications, and type and affix labels.
54| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 3/5 | Large hospital and chain pharmacies have adopted robotic dispensing systems, but small independent pharmacies and rural locations lag; adoption is steady but not rapid because capital costs and integration complexity create friction despite clear ROI. |
| Sector adoption velocity | claude-sonnet-5 | 3/5 | Automated dispensing and packaging systems have been adopted meaningfully in large hospital and retail pharmacy chains, but adoption is uneven and many independent pharmacies still use manual processes. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI-assisted systems augment technicians by automating routine fills and labeling, allowing technicians to focus on exception handling, clinical checking, and patient interaction, thereby raising their value and productivity in a complementary workflow. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI and automation tools significantly speed up label generation, inventory-linked packaging, and error-checking, meaningfully boosting technician throughput while humans retain oversight and physical handling. |
| Task automatability | claude-haiku-4-5-20251001 | 4/5 | Modern robotic pharmacy systems can handle bulk unpacking, dispensing precise doses into bottles, and applying labels with high accuracy and speed. While some exceptions and manual verification remain necessary, the core workflow—unpacking, filling, and labeling—can achieve >50% time savings at equal or better quality in controlled pharmacy environments. |
| Task automatability | claude-sonnet-5 | 3/5 | The physical filling/prepacking requires robotic hardware, but label typing and generation is easily automated; overall the task is a mix of easily-automatable data entry and physical manipulation that current general AI cannot fully perform without specialized robotics.' |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | Regulatory oversight (FDA, state pharmacy boards) and liability for dose accuracy create moderate friction, and human verification of high-risk dispensing is often mandated or preferred; however, no per-task licensing requirement prevents automation, and pharmacist sign-off (not technician manual work) is the legal gate. |
| Adoption barriers | claude-sonnet-5 | 4/5 | Pharmacy operations are heavily regulated, with legal requirements for licensed oversight of dispensing accuracy and labeling, creating strong regulatory and liability barriers to full automation without human check. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 4/5 | Initial capital investment is high, but per-dose labor cost of robotic dispensing (amortized) is substantially cheaper than manual technician labor once deployed; ongoing maintenance and system integration costs are moderate relative to saved technician hours. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | Robotic pharmacy automation systems require significant capital investment, integration, and maintenance, so while cheaper per-unit at high volume, the all-in cost is not clearly an order of magnitude below technician wages for smaller-scale operations. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 4/5 | Deployed automated dispensing systems (carousel-based, robotic arms, and countertop devices) are in production use across hospital and retail pharmacies, reliably performing bulk prepacking and labeling at scale, though integration with legacy systems and exception handling still require technician oversight. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | Pharmacy automation systems (robotic dispensing, packaging machines) are deployed in many large pharmacies and hospitals today, but many smaller pharmacies still rely on manual prepacking and labeling, so reliability varies by setting. |
Assist customers by answering simple questions, locating items, or referring them to the pharmacist for medication information.
28CI 25–30 · exposure 25 · augmentation 50 · importance 4.5/5 · click for rater detail
Assist customers by answering simple questions, locating items, or referring them to the pharmacist for medication information.
28| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | While some large chains have trialed self-service kiosks and chatbots, mainstream adoption remains limited; most independent and mid-size pharmacies lack digital infrastructure, and regulatory caution about medication information keeps human technicians as the standard. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Retail pharmacy is a moderately digitized sector with slow, uneven adoption of AI for in-store customer interactions compared to leading digital sectors. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI could assist by suggesting inventory locations, drafting responses to simple questions, or flagging potential escalation scenarios, moderately raising technician productivity without replacing the human role in judgment and customer interaction. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI tools like chatbots, inventory lookup systems, and store apps can help technicians quickly answer questions or find item locations, offering moderate productivity gains. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI could answer simple, factual questions and locate inventory items with setup, the task requires judgment about when to escalate to a pharmacist and navigation of a physical store environment, which current systems cannot do end-to-end reliably enough to meet the 50% time-saving threshold at equal quality. |
| Task automatability | claude-sonnet-5 | 2/5 | Answering simple questions and directing customers could partially be automated via kiosks or chatbots, but locating items and in-person interaction require physical presence and situational judgment that current AI cannot fully replace end-to-end.11 The referral logic is simple but the physical retail interaction limits full automation. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Strong barriers exist: customers expect human assistance for medication-related advice, liability concerns deter full automation of referral decisions, and regulations may require human oversight of pharmacy operations; organizational friction and customer preference for in-person service create additional protection. |
| Adoption barriers | claude-sonnet-5 | 3/5 | Medication-related questions must be referred to a licensed pharmacist, creating a regulatory carve-out, though the general customer service portion has fewer legal barriers. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | An AI chatbot or kiosk requires significant upfront integration, infrastructure, and ongoing maintenance; the cost per interaction is comparable to or higher than a technician's hourly wage for simple customer service tasks, especially accounting for oversight and system management. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | Deploying kiosks or voice assistants in physical retail requires hardware, integration, and maintenance costs that are not clearly cheaper than a low-wage technician performing this task. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Chatbots exist for basic pharmacy Q&A, but deployed systems have limited ability to locate physical items in-store, understand nuanced customer needs, or judge escalation correctly; they operate in narrow digital channels, not as in-store assistants. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Some pharmacies use chatbots or self-service kiosks for basic FAQs, but no widely deployed product reliably handles in-store customer assistance including physical item location at scale. |
Maintain and merchandise home healthcare products or services.
24CI 14–35 · exposure 20 · augmentation 38 · importance 3.6/5 · click for rater detail
Maintain and merchandise home healthcare products or services.
24| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Pharmacy remains a traditionally regulated, human-centric sector with slower digital transformation compared to e-commerce or finance. Adoption of automation for merchandising and maintenance is limited, with most pharmacies relying on manual technician work. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Retail pharmacy floor operations and physical merchandising show minimal AI adoption; this is a low-digitization, physically-grounded task. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI-driven inventory management systems and demand forecasting tools can meaningfully assist technicians in planning and organizing stock, reducing time spent on manual counting and ordering. However, the physical arrangement and customer-facing aspects limit the scope of augmentation. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI can assist with inventory analytics or planogram suggestions, but offers limited direct help with the physical merchandising and maintenance work itself. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | Merchandising and maintaining physical home healthcare products requires spatial reasoning, inventory management, and customer-facing decisions that AI cannot fully automate today. AI can assist with inventory tracking and ordering, but the physical merchandising and service coordination aspects remain heavily manual. |
| Task automatability | claude-sonnet-5 | 2/5 | This is a physical merchandising and inventory task requiring in-store product placement, stocking, and display management, which current AI cannot perform end-to-end.' |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Pharmacy operations are heavily regulated by state pharmacy boards and the DEA; merchandise standards and product safety requirements create compliance barriers. Additionally, customer preference for human interaction in healthcare settings and legal liability for product placement/maintenance create meaningful adoption friction. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement for merchandising itself, though it's bundled with pharmacy technician roles that have some organizational and physical-presence constraints. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | The loaded cost of AI systems for inventory management and coordination, plus integration and ongoing human oversight for physical merchandising tasks, remains comparable to or exceeds the cost of a pharmacy technician performing these duties. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | Physical stocking and merchandising still requires human labor or robotics far more costly than a technician's wage for this task; no cheap AI substitute exists. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | While inventory management systems exist, end-to-end autonomous maintenance and merchandising of physical products at retail/pharmacy locations requires handling, display arrangement, and quality checks that current AI systems cannot reliably perform without significant human oversight and physical robotics. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed AI product autonomously merchandises physical retail shelves or manages home healthcare product displays in pharmacies today. |
Maintain proper storage and security conditions for drugs.
23CI 20–25 · exposure 25 · augmentation 63 · importance 4.7/5 · click for rater detail
Maintain proper storage and security conditions for drugs.
23| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Large hospital and chain pharmacies have adopted automated storage systems, but small independent pharmacies and retail chains lag; overall adoption remains uneven and slow due to high capital requirements, regulatory friction, and entrenched labor practices. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Pharmacy settings adopt automation slowly for physical security tasks; while automated dispensing cabinets are common, full security/monitoring automation is still emerging and unevenly deployed. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI-assisted monitoring systems that alert technicians to temperature excursions, expired stock, or inventory discrepancies significantly enhance a technician's ability to maintain compliance and catch problems faster, transforming workflow efficiency while keeping the human in supervisory control. |
| Augmentation potential | claude-sonnet-5 | 3/5 | Smart sensors, automated alerts, and inventory management software help technicians monitor conditions more efficiently, though the core security and compliance judgment remains human-led. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | Maintaining storage temperature, humidity, and inventory systems can be partially automated via sensors and robotics, but physical inspection, security protocols, and intervention for anomalies require human judgment and presence that current AI cannot fully replicate end-to-end at the required quality threshold. |
| Task automatability | claude-sonnet-5 | 2/5 | This is largely a physical task involving monitoring storage conditions (temperature, controlled substance security, shelf organization) that requires physical presence and manual verification, limiting end-to-end AI automation.' |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Strong regulatory and legal barriers exist: pharmacy technicians must be licensed, DEA controls require documented human accountability for controlled substances, and chain-of-custody regulations mandate human verification and sign-off on storage security protocols that autonomous systems cannot legally satisfy. |
| Adoption barriers | claude-sonnet-5 | 4/5 | Controlled substance storage and security is heavily regulated (DEA, state pharmacy boards) requiring licensed/registered personnel accountability, creating strong regulatory and liability barriers to full automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Automated inventory and monitoring systems have high upfront capital costs and ongoing maintenance; for small to mid-size pharmacies, the total cost of ownership for automation can exceed the loaded wage of technicians, though large chains may achieve better ratios. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | Sensor-based monitoring systems add cost on top of still-needed human labor for physical security and compliance, so AI does not clearly reduce cost versus a human doing the full task. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Automated storage and retrieval systems exist in some large pharmacies, and environmental monitoring is deployed, but comprehensive autonomous management of drug storage conditions combined with security oversight remains limited; most real-world systems require human pharmacy technicians for final verification and intervention. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | IoT sensors and monitoring systems exist for temperature/humidity tracking and inventory alerts, but the physical security, controlled substance handling, and compliance verification still require human technicians on-site. |
Deliver medications or pharmaceutical supplies to patients, nursing stations, or surgery.
21CI 16–25 · exposure 17 · augmentation 25 · importance 4.4/5 · click for rater detail
Deliver medications or pharmaceutical supplies to patients, nursing stations, or surgery.
21| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Adoption of autonomous delivery in hospitals is slow and concentrated in large academic medical centers. Most community pharmacies and smaller healthcare settings continue manual delivery by pharmacy technicians, reflecting capital constraints, regulatory caution, and organizational friction. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Healthcare logistics automation (delivery robots, tube systems) is adopted unevenly and mainly in large hospital systems, with slow diffusion due to capital costs and regulatory caution. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | AI offers minimal augmentation for a primarily physical logistical task. Mobile carts with dispatch optimization assist routing, but the core act of delivering to multiple locations remains human-driven; automation does not meaningfully amplify human judgment or decision-making in the task itself. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI can assist with tracking, routing, or inventory alerts for deliveries, but does not substantially transform the physical act of delivering medications. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | Physical delivery to multiple locations requires navigation, handling, and person-to-person handoff in real-world environments. While mobile robotics exist, they cannot yet reliably manage complex hospital layouts, stairs, security protocols, and human interaction at scale, and even specialized delivery bots in controlled hospital settings represent partial automation requiring significant supervision. |
| Task automatability | claude-sonnet-5 | 1/5 | This is a physical delivery task requiring transport of medications to patients, nursing stations, or surgery locations, which current AI systems cannot perform end-to-end without robotic hardware.' |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Delivery of controlled substances (opioids, etc.) requires chain-of-custody compliance and often licensed pharmacy oversight. Hospitals have liability and regulatory concerns around autonomous systems handling patient safety-critical deliveries, and many facilities prefer human accountability for medication transport. |
| Adoption barriers | claude-sonnet-5 | 4/5 | Medication delivery involves chain-of-custody, verification, and safety protocols often requiring a licensed or supervised human handler, plus liability concerns around misdelivery of controlled substances. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Specialty hospital delivery robots (e.g., TUG, Relay) cost $100k–$300k+ per unit, require installation and maintenance, and handle only simplified routes. The loaded cost per delivery typically exceeds a pharmacy technician's hourly wage, especially when accounting for infrastructure and oversight. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | Robotic delivery systems require significant capital investment in hardware and facility infrastructure, making them costly relative to human technicians for most settings, though large hospitals may see savings at scale. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Autonomous delivery robots are deployed in narrow, controlled hospital settings (some institutions use AMRs for medication runs) but face material reliability issues with real-world complexity, human handoffs, and exception handling. No general-purpose, off-the-shelf system reliably performs this end-to-end across diverse environments. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Some hospitals use automated delivery robots or pneumatic tube systems for pharmaceutical transport, but these are narrow, facility-specific deployments rather than general AI products performing the full task. |
Clean and help maintain equipment or work areas and sterilize glassware, according to prescribed methods.
14CI 5–23 · exposure 8 · augmentation 13 · importance 4.4/5 · click for rater detail
Clean and help maintain equipment or work areas and sterilize glassware, according to prescribed methods.
14| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Pharmacy automation has focused on dispensing and inventory management; cleaning and sterilization remain low-priority targets for investment. Physical task automation in healthcare settings is slow and limited to large institutional chains, not mainstream adoption. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Physical cleaning and sterilization tasks in healthcare settings show minimal AI/robotic adoption; this remains a manual task with low digitization relevance. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | AI could assist in scheduling sterilization runs or monitoring compliance records, but the core manual and sensory work of cleaning and handling equipment offers limited augmentation value compared to direct task execution or human oversight. |
| Augmentation potential | claude-sonnet-5 | 1/5 | AI offers no meaningful assistance for physical cleaning and sterilization procedures, as these require hands-on manual execution rather than information processing. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While equipment cleaning and sterilization involve procedural steps, they require physical manipulation of glassware and equipment in varied spatial configurations, detection of contamination, and real-time adjustments—capabilities that today's AI and robotics handle only in controlled, narrowly scoped settings. Partial automation of monitoring or scheduling is possible, but end-to-end autonomous execution at quality parity remains out of reach. |
| Task automatability | claude-sonnet-5 | 1/5 | Physical cleaning and sterilization of glassware and equipment requires manual dexterity and physical presence that current AI systems cannot perform.: |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Regulatory requirements (FDA, USP standards) mandate specific sterilization methods and validation procedures; pharmacies must document compliance with prescribed sterilization protocols, creating a strong requirement for traceable, human-validated processes rather than unsupervised automation. |
| Adoption barriers | claude-sonnet-5 | 3/5 | While not licensed-specific, sterilization protocols and safety/compliance standards in pharmacy settings create procedural friction, though the barrier is more about physical automation than regulatory licensing. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Robotic systems capable of manipulating fragile glassware and performing sterilization checks remain expensive to acquire, integrate, and maintain, making them more costly than hiring pharmacy technicians for these routine tasks in most settings. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no AI-based substitute for this physical labor, so AI cost comparison is not applicable; human labor remains the only viable option. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed commercial products reliably perform general cleaning, glassware handling, and sterilization validation in real pharmacy environments. Specialized cleaning robots exist for industrial contexts but are not routinely integrated into pharmacy workflows at scale. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed AI product performs physical cleaning or sterilization tasks in pharmacy settings; this remains purely a human/robotic-mechanical task outside AI's domain. |
Transfer medication from vials to the appropriate number of sterile, disposable syringes, using aseptic techniques.
13CI 0–25 · exposure 13 · augmentation 38 · importance 4.5/5 · click for rater detail
Transfer medication from vials to the appropriate number of sterile, disposable syringes, using aseptic techniques.
13| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Adoption remains slow outside large hospital systems and specialty compounding centers. Most community and small retail pharmacies continue manual transfer; even in hospitals, robotic systems are pilots or limited to high-volume generic operations, not mainstream production. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Pharmacy technician physical compounding work remains low-digitization and manual; adoption of automation here is confined to niche robotic dispensing systems in some hospitals, not general AI agents. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | Current robotic and vision-assisted tools can help technicians by automating high-volume repetitive transfers or flagging vial/syringe mismatches, but the technician remains responsible for aseptic technique verification and final quality checks. Modest productivity gains are possible, but the human remains central to compliance and safety. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI can assist with related administrative tasks like dosage calculations or inventory tracking, but offers minimal direct assistance to the physical aseptic transfer procedure itself. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While robotic systems exist for medication transfer in some hospital settings, they require significant setup, calibration, and human oversight. Current AI-based vision and manipulation systems cannot reliably handle the full sterile protocol, syringe variability, and real-time aseptic technique compliance needed to achieve 50% time savings at equal quality without extensive human intervention. |
| Task automatability | claude-sonnet-5 | 1/5 | This is a physical manual manipulation task requiring sterile technique with syringes and vials; no off-the-shelf AI system can perform this physical compounding action end-to-end today. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Regulatory and professional barriers are substantial: aseptic technique compliance is legally mandated under pharmacy practice acts, USP <797> standards require human oversight, and liability for medication errors typically rests on the licensed pharmacist. Hospitals and pharmacies face legal and safety accountability that prevents full automation substitution. |
| Adoption barriers | claude-sonnet-5 | 5/5 | Sterile compounding is heavily regulated (USP 797/800, state pharmacy boards) requiring certified/licensed personnel and strict aseptic protocols, with high liability for contamination errors. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Robotic pharmacy systems carry high capital and maintenance costs ($500k–$1M+), plus ongoing support and integration expenses. For most pharmacy settings, the loaded human cost (wages, benefits, training) remains competitive or lower than the amortized cost of current automation technology. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no viable AI-driven substitute for this physical task, so any comparison favors the human technician; specialized robotic compounders are costly capital investments, not cheap AI inference. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Limited robotic pharmacy automation (e.g., carousel-based systems) handles some high-volume scenarios in large hospitals, but these are narrow applications requiring custom integration. No general-purpose AI system performs this task reliably in diverse pharmacy settings; most deployments remain human-dependent for quality assurance and aseptic oversight. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed AI product performs aseptic drug transfer into syringes; robotic compounding systems exist only as specialized capital equipment in limited pharmacy settings, not general AI products. |
Mix pharmaceutical preparations, according to written prescriptions.
13CI 0–25 · exposure 13 · augmentation 38 · importance 4.4/5 · click for rater detail
Mix pharmaceutical preparations, according to written prescriptions.
13| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Adoption is slow and concentrated in large hospital systems and major chains; small independent pharmacies and many retail locations still rely entirely on manual technicians. Regulatory friction, upfront capital, and integration complexity limit deployment despite clear demand. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Pharmacy compounding is a highly manual, physically regulated task with minimal AI/robotic adoption in production settings. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI and robotic systems can assist technicians by automating routine dispensing, reducing manual counting/measuring time, and flagging potential drug interactions or dosing errors; however, the human remains responsible for verification and complex custom preparations, providing moderate productivity gain rather than transformation. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI can assist with calculating dosages, checking interactions, or verifying prescriptions, but offers little help with the physical mixing process itself. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI-driven robotic systems can perform liquid/solid dosing with precision, end-to-end prescription-to-dispensed-medication automation requires integration of multiple steps (prescription parsing, inventory management, quality checks, physical manipulation). Current systems handle isolated components but struggle with the full workflow's variability and safety requirements at equal quality in real pharmacy environments. |
| Task automatability | claude-sonnet-5 | 1/5 | Physical compounding of pharmaceutical preparations requires manual dexterity, sterile technique, and hands-on manipulation of materials that current AI cannot perform without robotics far beyond typical deployment. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Strict FDA and state pharmacy board regulations mandate human verification of drug preparations, require pharmacist oversight, and impose liability standards that make full unsupervised automation legally infeasible. Error-cost asymmetry is severe (patient harm liability) and human sign-off is effectively mandated by law. |
| Adoption barriers | claude-sonnet-5 | 5/5 | Compounding is tightly regulated, requires licensed/certified personnel, pharmacist verification, and carries high liability for errors, making substitution legally and practically barred. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Robotic systems for pharmacy mixing are capital-intensive ($500k–$2M+), require maintenance and integration, and still demand technician oversight. For many smaller or mid-size pharmacies, total cost of ownership exceeds multiple technician salaries; only high-volume settings achieve favorable cost ratios. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no viable AI substitute for the physical act of compounding, so cost comparison favors the human technician entirely. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Robotic pharmacy dispensing systems exist and operate in some hospital and chain pharmacies, but they remain narrow in scope (e.g., handling high-volume standard pills), have measurable error rates, and require extensive human oversight and manual handling of complex/custom preparations. No system reliably replaces the technician end-to-end in production. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed AI product physically mixes pharmaceutical preparations; this remains a manual task performed by trained technicians under pharmacist supervision. |
Restock intravenous (IV) supplies and add measured drugs or nutrients to IV solutions under sterile conditions to prepare IV packs for various uses, such as chemotherapy medication.
6CI 0–11 · exposure 5 · augmentation 38 · importance 4.4/5 · click for rater detail
Restock intravenous (IV) supplies and add measured drugs or nutrients to IV solutions under sterile conditions to prepare IV packs for various uses, such as chemotherapy medication.
6| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Despite automation interest in pharmacy, actual adoption of IV compounding robots remains slow and concentrated in large hospital systems. Most pharmacies still rely on technician-led sterile compounding; market penetration is limited by cost, regulatory complexity, and institutional resistance to delegating chemotherapy prep fully to machines. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Healthcare pharmacy compounding is a highly regulated, physically-grounded sector with low AI adoption for hands-on sterile preparation tasks; robotic compounding exists but adoption is slow and limited to large institutions. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI-assisted inventory management, barcode verification, and dose calculation tools can help technicians work faster and reduce errors in some preparatory steps. However, the core tasks of sterile handling and visual inspection remain human-dependent, limiting the augmentation upside. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI/software can assist with inventory tracking, dosage calculations, or barcode verification to reduce errors, but it does not meaningfully augment the physical sterile compounding process itself. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | This task requires precise physical manipulation in sterile environments, handling hazardous chemotherapy agents, and real-time visual inspection of solutions and vials. Current AI/robotic systems lack the dexterous manipulation, safety protocols, and adaptive decision-making needed for reliable end-to-end execution of IV pack preparation. |
| Task automatability | claude-sonnet-5 | 1/5 | This is a physical sterile compounding and manual restocking task requiring hand dexterity and aseptic technique; no current AI system can physically prepare IV admixtures. |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Pharmacy compounding of chemotherapy is heavily regulated by state pharmacy boards and federal guidelines (USP <797>, <825>); preparation often requires licensed pharmacist verification and sign-off. Liability for adverse patient outcomes, hazardous drug handling protocols, and sterile compounding certification create hard legal and professional barriers to full automation. |
| Adoption barriers | claude-sonnet-5 | 5/5 | Sterile compounding of chemotherapy and IV drugs is heavily regulated (USP <797>/<800>, state pharmacy boards) and requires certified/licensed personnel under strict oversight, creating hard legal and safety barriers to substitution. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Robotic IV compounding equipment is capital-intensive and requires substantial facility modifications, training, and maintenance. Integration costs and the need for human verification of sterile compounding make the total cost-per-task comparable to or higher than trained technician wages in most settings. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | AI cannot perform the physical preparation, so there is no viable AI cost comparison; any automation would require specialized robotic hardware (e.g., IV compounding robots), which is capital-intensive, not a cheap AI substitute. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | While some pharmacy automation systems exist for counting and labeling, they do not reliably perform the complete sterile compounding, drug measurement, and solution preparation for chemotherapy IVs. Clinical deployment remains limited and typically requires significant human oversight and final verification. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed AI product performs sterile IV compounding or physical restocking; this remains a manual clinical task performed by trained humans, sometimes with robotic compounders but not general AI systems. |
Related occupations — Healthcare Practitioners & Technical
How to read this
A high substitution score does not mean this job disappears — it means a large share of its current tasks face replacement pressure, so the mix of tasks is likely to change. High augmentation alongside substitution typically means the occupation reorganizes around the protected tasks. Wide confidence intervals mean the rater panel disagreed: treat those scores as open questions, not verdicts.
What would change this score
New model capabilities (automatability, feasibility), falling inference costs (cost ratio), regulation and licensing shifts (barriers), and measured sector adoption (velocity) all re-enter at every index release. Each release is recomputed, versioned and kept queryable — scores are claims with a date on them, not permanent labels.