Postal Service Mail Carriers
43-5052.00Median wage $60,550/yr328,820 employed (US)Rank #255 of 923 scored · top 28% by substitution
Sort and deliver mail for the United States Postal Service (USPS). Deliver mail on established route by vehicle or on foot. Includes postal service mail carriers employed by USPS contractors.
Sub-scores
0–100 · band = confidence interval from rater disagreement
Substitution35
Exposure34
Augmentation37
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
14%
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%34
panel mean rating 2.4/5 → substitution pressure 34/100
Technical feasibility todayw 20%32
panel mean rating 2.3/5 → substitution pressure 32/100
Cost vs. human wagew 15%33
panel mean rating 2.3/5 → substitution pressure 33/100
Adoption barriersw 20%inverted — strong barriers lower the score47
panel mean rating 3.1/5 (barrier strength) → substitution pressure 47/100
Sector adoption velocityw 10%25
panel mean rating 2.0/5 → substitution pressure 25/100
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.
Maintain accurate records of deliveries.
85CI 75–95 · exposure 87 · augmentation 75 · importance 4.3/5 · click for rater detail
Maintain accurate records of deliveries.
85| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 5/5 | Major postal and logistics operators have deployed automated delivery tracking, scanning, and record systems industry-wide for over a decade. Adoption is near-universal in information-intensive logistics sectors. |
| Sector adoption velocity | claude-sonnet-5 | 4/5 | Postal and logistics industries have already broadly adopted electronic scanning and tracking systems, representing a mature, high-penetration adoption pattern. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | Current systems already augment carrier productivity by automatically logging deliveries as they scan packages, reducing manual paperwork and enabling real-time tracking visibility for both carrier and customer. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI-enabled scanning, route optimization, and automatic logging significantly reduce manual record-keeping burden while carriers still perform physical delivery and exception handling. |
| Task automatability | claude-haiku-4-5-20251001 | 5/5 | Delivery record maintenance is largely data entry and structured logging—scanning barcodes, updating delivery status in databases, timestamping events. Current AI and postal automation systems already handle this end-to-end with well over 50% time savings via automated scanning, GPS tracking, and database population. |
| Task automatability | claude-sonnet-5 | 4/5 | Recording delivery data is largely digital scanning/logging already integrated into handheld devices, and software can automatically log, timestamp, and reconcile delivery records with minimal human input. |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | There are minimal legal or regulatory barriers to automating delivery logging itself; the main friction is that carriers still require a human to physically complete the delivery (the primary task). Once the delivery occurs, recording it is routine. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement for record-keeping itself, though union rules, data accuracy liability, and organizational integration with existing carrier workflows create some friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | Automated barcode scanning, GPS logging, and database updates cost pennies per delivery once infrastructure is amortized, whereas manual record-keeping by a human requires wages and overhead. The AI cost is orders of magnitude lower. |
| Cost vs. human wage | claude-sonnet-5 | 4/5 | Automated scanning and logging systems are cheap per-package compared to manual record-keeping labor, though hardware and integration costs exist. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 5/5 | Postal carriers today use handheld devices and integrated tracking systems that automatically log deliveries, confirmations, and exceptions. These are mature, deployed products in production across USPS, UPS, FedEx, and other carriers at scale. |
| Technical feasibility today | claude-sonnet-5 | 4/5 | Barcode scanners, GPS tracking, and delivery management software (used by USPS, UPS, FedEx) already reliably automate delivery record-keeping in production at scale. |
Scan labels on letters or parcels to confirm receipt.
82CI 69–95 · exposure 87 · augmentation 50 · importance 4.8/5 · click for rater detail
Scan labels on letters or parcels to confirm receipt.
82| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 5/5 | Postal and logistics sectors have rapidly adopted automated scanning systems over the past 15+ years; this is among the most widely deployed automations in the industry globally. |
| Sector adoption velocity | claude-sonnet-5 | 3/5 | Postal and logistics sectors have adopted scanning technology broadly but are not fast-moving on further AI innovation beyond existing barcode systems; robotics/autonomous delivery adoption remains slow.3 |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI-driven scanning can flag anomalies (damaged packages, unreadable labels) to assist human workers in exception handling, improving their efficiency on edge cases that remain in their workflow. |
| Augmentation potential | claude-sonnet-5 | 3/5 | Scanners assist carriers by automating proof-of-delivery and reducing paperwork, offering clear but incremental productivity gains rather than transformative change.3 |
| Task automatability | claude-haiku-4-5-20251001 | 5/5 | Scanning labels on letters or parcels is a purely visual, mechanical task that current barcode and QR-code reading systems (including mobile CV models) perform reliably end-to-end with near-instantaneous processing, easily exceeding 50% time savings over manual confirmation. |
| Task automatability | claude-sonnet-5 | 4/5 | Scanning barcodes/labels is a simple, well-defined physical-digital action already largely automated via handheld scanners; only the physical act of moving to the scan point remains human, but the recognition/confirmation logic is fully machine-executable.4 |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | While some postal regulations may prefer human verification and customer-facing touchpoints exist, the core scanning task itself faces minimal legal or licensing barriers; organizations can and do automate it with standard IT infrastructure. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing or legal requirement mandates a human scan the label; the main barrier is that scanning is bundled with physical delivery which still requires a person present.2 |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | Barcode scanning via mobile or stationary systems costs pennies per scan in deployment and operation, orders of magnitude cheaper than paying a human postal worker to visually verify and manually log each item. |
| Cost vs. human wage | claude-sonnet-5 | 3/5 | The scanning hardware/software itself is cheap, but it still requires a human carrier to physically carry and operate the device during delivery, so overall cost savings versus the human labor are moderate, not transformative.3 |
| Technical feasibility today | claude-haiku-4-5-20251001 | 5/5 | Deployed barcode/QR scanning products are mature and widely used in logistics and postal operations at scale; postal services globally already use handheld scanners and mobile apps to confirm receipt of items automatically. |
| Technical feasibility today | claude-sonnet-5 | 4/5 | Postal services worldwide already deploy handheld barcode/RFID scanners and route-tracking devices that reliably log delivery confirmation at scale in production.4 |
Record address changes and redirect mail for those addresses.
76CI 72–79 · exposure 75 · augmentation 50 · importance 4.1/5 · click for rater detail
Record address changes and redirect mail for those addresses.
76| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 4/5 | Postal services and logistics have been among the earliest and deepest adopters of automation and AI-driven systems for routing, address validation, and data management. Major carriers and USPS have deployed automated address-change processing and mail-forwarding systems for years, representing mature adoption in a digitized sector. |
| Sector adoption velocity | claude-sonnet-5 | 3/5 | Postal services have adopted automation for sorting and address processing over years, but overall postal sector digitization and AI adoption velocity is moderate compared to fully digital industries. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI assists postal workers by auto-populating address fields, flagging potential duplicates or errors, and pre-sorting redirection requests, raising their speed and accuracy. However, the augmentation is incremental; human judgment remains useful for exception handling but is not transformative to core task performance. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI-assisted systems help human staff verify and manage address change requests and exceptions, improving speed and accuracy while humans still oversee edge cases and customer service issues. |
| Task automatability | claude-haiku-4-5-20251001 | 4/5 | Address change recording and mail redirection are highly structured data-entry and routing tasks. Current AI systems can reliably parse address information, validate formats, update database records, and trigger mail forwarding instructions with minimal human oversight, achieving well over 50% time savings at comparable quality to manual entry. |
| Task automatability | claude-sonnet-5 | 4/5 | Recording address changes and generating redirect instructions is a structured data-entry and rules-based task that AI/automation systems can handle with high accuracy and significant time savings once integrated with postal databases. |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | While USPS faces some organizational friction around legacy systems integration and the need for human oversight of complex cases, there are no hard regulatory barriers that mandate a licensed human perform address updates or mail redirection. Customer preference for human contact is minimal for this administrative task. |
| Adoption barriers | claude-sonnet-5 | 2/5 | Some verification and fraud-prevention requirements exist (identity confirmation for address changes) but these are largely already handled by automated systems, not requiring licensed human judgment. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | Automating address recording and redirection via database systems costs far less per transaction than a postal worker manually entering data and managing routing slips. The cost per address processed by AI is orders of magnitude lower than loaded wages for equivalent manual output. |
| Cost vs. human wage | claude-sonnet-5 | 4/5 | Automated address-change processing and mail-redirect systems cost far less per transaction than manual clerical handling, though integration with physical sorting still requires some infrastructure investment. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 4/5 | USPS and postal services already deploy address verification and mail-forwarding automation systems in production. While human review of edge cases (unusual address formats, ambiguous customer requests) still occurs, mature APIs and workflow systems reliably handle the majority of routine address changes and redirection at scale. |
| Technical feasibility today | claude-sonnet-5 | 4/5 | USPS and other postal services already use automated systems (e.g., online change-of-address forms, NCOA databases, sorting automation) to process and apply forwarding instructions at scale in production. |
Complete forms that notify publishers of address changes.
69CI 65–72 · exposure 70 · augmentation 50 · importance 3.6/5 · click for rater detail
Complete forms that notify publishers of address changes.
69| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 3/5 | The Postal Service has digitized significant back-office work but is a large, legacy-heavy organization with slow technology rollout. While the technical feasibility is high, actual deployment of end-to-end automation for publisher notifications has been uneven; pilots exist but full fleet adoption lags behind private-sector information services. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Postal service operations are a slow-adopting, heavily unionized, government-regulated sector with limited pace of AI integration into physical/clerical workflows. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can assist mail carriers by pre-filling forms, suggesting matching publishers, and flagging unusual address patterns, reducing manual data entry and lookup time. However, the task is already fairly routine, so augmentation value is moderate compared to higher-judgment tasks. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI-assisted address validation and auto-fill tools can speed up completion of these forms, though the task is simple enough that gains are moderate. |
| Task automatability | claude-haiku-4-5-20251001 | 4/5 | Notifying publishers of address changes is largely a data-entry and form-completion task with structured, repetitive fields. Current AI systems can extract address information, populate standardized forms, and generate notifications with high accuracy, achieving significant time savings. Some human judgment around edge cases or non-standard address formats may still be needed, but the core task is highly automatable. |
| Task automatability | claude-sonnet-5 | 4/5 | This is a routine, structured data-entry task (matching address changes to publisher notification forms) well within current AI/automation capability, especially with digital forms and address databases.br |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | Few regulatory or licensing barriers exist for automating form completion and notifications; the Postal Service operates these internally and controls the process. Customer authorization is already implicit in address-change requests, and there is no legal requirement that a human personally complete each form. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement for this clerical subtask, though USPS internal systems and processes create some organizational integration friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 4/5 | The cost of AI automation (form-filling, notification dispatch, minimal human oversight) is substantially lower than paying a mail carrier to manually complete and route dozens of individual publisher notification forms per day. Inference and integration costs are negligible relative to hourly labor. |
| Cost vs. human wage | claude-sonnet-5 | 4/5 | Automated form-filling and address matching software is far cheaper per transaction than a human manually completing and forwarding these forms. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 4/5 | Deployed products for form automation, mail data extraction, and API-based notification systems already handle similar workflows in production environments. Postal service software and third-party address-change platforms can reliably complete and route these notifications at scale, though integration with specific publisher systems may require some customization. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | Automated address-change notification systems exist in mail processing (e.g., NCOA-style systems), but many carriers still handle this manually as part of paper-based route paperwork, so deployed reliability varies by context. |
Sort mail for delivery, arranging it in delivery sequence.
59CI 35–84 · exposure 59 · augmentation 50 · importance 4.6/5 · click for rater detail
Sort mail for delivery, arranging it in delivery sequence.
59| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | USPS and other postal carriers have adopted sorting machines incrementally since the 1990s but adoption remains slow in smaller facilities and rural areas due to cost and union constraints. Most mail still involves human sorting steps; the sector lags digitization-leading industries in AI adoption velocity. |
| Sector adoption velocity | claude-sonnet-5 | 4/5 | Postal services worldwide have already deeply adopted automated sequencing technology over recent decades, though final integration with human carriers remains standard practice. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | Barcode scanning, OCR assistance, and address validation tools already help mail carriers by highlighting address ambiguities and organizing bundles. These systems reduce manual lookup time and errors, providing useful but not transformative assistance—the human carrier remains essential for judgment and final delivery. |
| Augmentation potential | claude-sonnet-5 | 3/5 | Where automated sequencing isn't fully implemented, sorting software and handheld scanning tools assist carriers in organizing routes, offering moderate productivity gains. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | Postal sorting involves both structured machine-readable sorting (barcodes, ZIP codes) and unstructured handwriting recognition on variable mail formats. While OCR and barcode reading are mature, end-to-end sorting to delivery sequence still requires handling edge cases (illegible addresses, irregular packages) and real-world variation that current AI cannot reliably automate at 50% time savings without significant human oversight. |
| Task automatability | claude-sonnet-5 | 4/5 | Mail sequencing is a highly structured, rules-based sorting task well-suited to automated systems; USPS already uses automated sorting equipment extensively for this purpose. |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | Postal operations are government-regulated (USPS) with union contracts and service-level obligations that create friction toward full automation. Labor agreements, requirement to maintain geographic coverage, and liability for misdelivery introduce organizational and contractual barriers, though not absolute legal restrictions on automation. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing or legal requirement mandates human sorting; some organizational friction exists from unionized labor and legacy infrastructure but no hard legal barrier. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Sorting machines (AFSM, delivery sequencing systems) require substantial capital investment and ongoing maintenance; inference costs are low but infrastructure costs remain high relative to a mail carrier's marginal sorting labor. Humans sorting is still often cheaper per piece for small/mixed volumes, particularly in rural routes. |
| Cost vs. human wage | claude-sonnet-5 | 5/5 | Automated sorting machines process thousands of pieces per hour at a fraction of the labor cost of manual sorting, making automation vastly cheaper per unit. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 3/5 | Postal services deploy optical scanning and automated sorting machines for standardized mail, but these systems handle only pre-sorted or clearly formatted items. Products exist (USPS AFSM systems, barcode readers) but require clean inputs and human intervention for problem mail, so deployment is narrow and material error rates persist on diverse mail streams. |
| Technical feasibility today | claude-sonnet-5 | 5/5 | Delivery Bar Code Sorters and Delivery Point Sequencing machines have been in production use by postal services for decades, reliably automating this exact task at scale. |
Answer customers' questions about postal services and regulations.
57CI 52–61 · exposure 50 · augmentation 63 · importance 4.0/5 · click for rater detail
Answer customers' questions about postal services and regulations.
57| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 4/5 | Postal services and logistics companies have rapidly deployed AI chatbots and automated response systems for customer inquiries, with measurable displacement of human-handled contacts in production environments across the USPS and private carriers. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Postal services are historically slow-moving, unionized, and physically-oriented organizations with limited AI deployment velocity for customer-facing carrier interactions, though online/app-based support channels are growing. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI can substantially augment mail carriers by providing instant access to curated postal regulations, rate lookups, and policy summaries, allowing them to answer customer questions more quickly and accurately while remaining in the loop. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI-powered apps and chatbots can help carriers and customers get quick answers to routine regulation questions, supplementing but not replacing the carrier's direct customer interaction role. |
| Task automatability | claude-haiku-4-5-20251001 | 3/5 | AI can handle approximately 50% of routine customer inquiries about standard postal services, rates, and basic regulations through chatbots, but struggles with edge cases, policy exceptions, and nuanced regulation interpretation that require human judgment and current customer context. |
| Task automatability | claude-sonnet-5 | 3/5 | Answering standard postal service questions is well-suited to chatbots and knowledge-base lookups, but many interactions occur in person during delivery and involve ad hoc, situational context AI cannot fully replace end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | Limited barriers exist; while some organizations prefer human contact for customer trust, there are no legal or licensing requirements mandating human staff answer postal questions, and regulatory oversight is light. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement to answer postal questions, but unionized government workforce, customer service expectations, and the fact this is embedded in a broader physical job create some organizational friction against separating and automating just this task. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 4/5 | AI systems for customer service have very low marginal cost per interaction once deployed, making them substantially cheaper than human mail carriers handling the same volume of inquiries when labor costs are factored in. |
| Cost vs. human wage | claude-sonnet-5 | 3/5 | A chatbot answering routine postal questions is cheap, but replacing the in-person, real-time carrier interaction requires additional infrastructure (kiosks, apps) making blended cost roughly comparable to the marginal cost of the carrier already performing other duties. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 3/5 | Chatbots and virtual assistants for postal services exist and handle straightforward FAQ-style questions reliably, but deployed systems have material error rates on complex or unusual inquiries and narrower scope than full human capability. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | Postal services and retailers already deploy chatbots/FAQ systems for shipping and regulation questions, but carrier-level in-person Q&A remains handled by humans with no deployed product substituting that specific interaction. |
Register, certify, and insure parcels and letters.
51CI 30–72 · exposure 50 · augmentation 63 · importance 4.5/5 · click for rater detail
Register, certify, and insure parcels and letters.
51| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 5/5 | Postal and parcel services are highly digitized and profit-driven. Large carriers (USPS, FedEx, UPS) have systematically rolled out automated registration, barcode tracking, and insurance systems across distribution networks. Adoption is deep, production-grade, and decades-long in the major carriers. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Postal services are a traditionally slow-adopting, unionized, government-linked sector with physical logistics constraints, showing only incremental AI adoption via self-service kiosks and automated sorting. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | Even where humans remain in the loop, AI assistance dramatically raises carrier productivity: auto-populated forms, real-time insurance quote calculations, weight/size verification prompts, and exception flagging allow carriers to process parcels much faster while reducing errors. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI-powered systems (e.g., automated address verification, digital tracking, kiosk-based insurance calculators) meaningfully speed up the paperwork and verification portions of this task even though physical handling remains human-driven. |
| Task automatability | claude-haiku-4-5-20251001 | 4/5 | Most of this task—data entry, label generation, insurance calculation, and certification logging—can be automated with current systems. Barcode scanning, weight/dimension capture, and database lookups are well-established. The main friction is occasional edge cases (disputed values, special instructions) and human verification for high-value items, but 50% time-saving at equal quality is readily achievable. |
| Task automatability | claude-sonnet-5 | 2/5 | This involves physical handling of mail items, verifying customer information, applying stamps/labels, and processing transactions at a counter or on a route, which requires physical presence and manipulation that current AI cannot perform end-to-end.the informational/data-entry portion is automatable but the physical handling is not. |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | Postal services are federally regulated (USPS in the U.S.), but registration and certification automation is already permitted and widely deployed. Liability for misclassification or underinsurance creates some oversight friction, and customers may request manual handling; however, no legal requirement mandates human sign-off for routine parcels. |
| Adoption barriers | claude-sonnet-5 | 3/5 | Postal services involve regulatory chain-of-custody requirements, proof of mailing/insurance liability, and in some cases legal certification standards that create moderate procedural barriers to full automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 4/5 | Automated systems (barcode scanners, thermal printers, database lookup) cost a few cents per parcel after amortization. The loaded wage for a mail carrier to manually register, certify, and insure a parcel is roughly $0.50–$1.50 per item. Automation is clearly cheaper even with integration and oversight overhead. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | While software for calculating postage/insurance is cheap, the physical task still requires a human carrier or postal worker, so overall cost savings from AI are limited to the administrative sliver of the task. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 4/5 | Postal services and shipping platforms (USPS, FedEx, UPS) have deployed automated parcel registration and certification systems in production for years. Weight/dimension sensors, label printers, and tracking databases work reliably at scale. Error rates are low for routine parcels; complex claims handling is the remaining manual bottleneck. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Self-service kiosks and online postage/insurance purchase systems exist and are deployed, but they don't fully replace carrier-handled registration/certification workflows which still require human verification and physical processing at delivery points. |
Meet schedules for the collection and return of mail.
44CI 10–77 · exposure 45 · augmentation 50 · importance 4.6/5 · click for rater detail
Meet schedules for the collection and return of mail.
44| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | The USPS and postal services globally remain primarily human-operated with limited public evidence of deep AI scheduling automation in production; while private couriers (FedEx, UPS) have adopted route optimization more aggressively, government postal agencies lag in digitization and deployment velocity. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Postal delivery is a highly physical, moderately digitized sector with minimal AI-driven route automation in production; adoption of autonomous delivery remains experimental and rare. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI systems can assist mail carriers in real-time by optimizing routes, flagging missed stops, predicting delivery timing, and identifying collection opportunities—enhancing human productivity significantly while the carrier remains responsible for physical execution and customer interaction. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI can help with route optimization and scheduling software, offering some indirect productivity gains, but it does not materially assist the physical act of collecting and returning mail. |
| Task automatability | claude-haiku-4-5-20251001 | 5/5 | Sorting and organizing mail routes to meet collection schedules is fundamentally a logistics optimization problem that can be fully automated with route-planning AI systems; collection itself involves physical pickup, but scheduling the collection rounds and optimizing return times can achieve >50% time savings with current AI and GPS routing technologies. |
| Task automatability | claude-sonnet-5 | 1/5 | This is a physical delivery/collection route task requiring driving/walking and time-managed handling of mail across locations, which current AI cannot perform end-to-end.dicionar Robots and autonomous vehicles are not deployed for this purpose at scale. |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | The U.S. Postal Service operates as a government entity with union agreements (NALC) that create friction around job displacement; there is no strict licensing barrier to automating scheduling itself, but organizational inertia, labor contracts, and the integration of automation into legacy postal systems create material adoption friction. |
| Adoption barriers | claude-sonnet-5 | 3/5 | While not licensed like a doctor, USPS carriers operate under civil service rules, union contracts, security/chain-of-custody requirements for mail, and physical route access constraints that create real organizational friction against automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | AI route optimization and scheduling incurs only marginal cloud compute and licensing costs per route, while a human mail carrier's fully loaded cost (salary, benefits, vehicle, oversight) is several tens of thousands annually; the cost difference is at least an order of magnitude in favor of automation. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no viable AI substitute for physical mail collection and delivery, so any comparison favors the human worker who can actually perform the task. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 4/5 | Deployed route optimization systems (Google Maps, delivery logistics platforms) already handle similar scheduling tasks at scale for courier and delivery services; mail-specific scheduling is narrowly deployed but the underlying technologies are mature and proven in production for analogous distribution logistics. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product performs the physical collection and return of mail on a schedule; this remains a human/physical logistics task not addressed by AI products. |
Travel to post offices to pick up the mail for routes or pick up mail from postal relay boxes.
42CI 10–74 · exposure 45 · augmentation 25 · importance 4.1/5 · click for rater detail
Travel to post offices to pick up the mail for routes or pick up mail from postal relay boxes.
42| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | While autonomous vehicles are advancing rapidly in tech and logistics sectors, postal services remain traditionally organized with slow organizational change; actual deployment of autonomous mail pickup is still in pilot phases rather than widespread production. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Postal and physical delivery sectors show minimal AI/autonomous vehicle adoption for this specific pickup task; the industry is a laggard in physical automation deployment. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | Current AI provides minimal assistance to mail carriers during the pickup task itself; GPS and route optimization offer marginal gains, but the human driver remains essential and AI does not substantially transform their productivity on this specific subtask. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI can assist with route optimization or scheduling of pickups, but offers little direct assistance to the physical act of traveling and retrieving mail. |
| Task automatability | claude-haiku-4-5-20251001 | 5/5 | Autonomous vehicle technology and robotics can fully replace the physical travel and mail pickup components with no human intervention, achieving well over 50% time savings while maintaining equal quality of service in controlled postal logistics networks. |
| Task automatability | claude-sonnet-5 | 1/5 | This requires physical transportation of mail via vehicle or on foot between locations, which no current AI system can perform end-to-end without robotic/autonomous vehicle hardware far beyond typical AI deployment. |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | Regulatory approval for autonomous vehicles on public roads, safety certification, and union labor agreements create meaningful friction, though no legal requirement mandates human presence for this specific task component. |
| Adoption barriers | claude-sonnet-5 | 3/5 | While not legally requiring a licensed human, postal regulations, security/chain-of-custody requirements for mail, and liability for lost mail create meaningful organizational and regulatory friction against automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 4/5 | Autonomous vehicle operation costs (fuel, maintenance, depreciation, oversight) are substantially lower than employing a human mail carrier when amortized per pickup cycle, though initial capital investment remains significant. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | Autonomous vehicle systems capable of this task would require expensive hardware, mapping, and safety oversight, making them far costlier than a human carrier's wage for this simple physical task. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 4/5 | Autonomous delivery vehicles and mail-sorting robots are in pilot and early production deployment by USPS and private carriers, though full end-to-end autonomous mail pickup at scale remains partially dependent on route optimization and facility coordination integration. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product autonomously travels to post offices and physically collects mail from relay boxes; autonomous delivery vehicles remain pilot/research stage in constrained settings. |
Provide customers with change of address cards and other forms.
36CI 0–72 · exposure 38 · augmentation 13 · importance 3.4/5 · click for rater detail
Provide customers with change of address cards and other forms.
36| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Postal service remains a traditional, heavily regulated sector with strong human-contact requirements. There is no meaningful AI adoption trajectory for physical form distribution at customer interfaces. |
| Sector adoption velocity | claude-sonnet-5 | 3/5 | USPS has moved much change-of-address processing online, but postal work overall is a slow-adopting, unionized, physically-oriented sector with mixed digitization. |
| Augmentation potential | claude-haiku-4-5-20251001 | 1/5 | AI provides no meaningful assistance to a mail carrier in the act of providing physical forms and cards to customers; the task is fundamentally manual and interpersonal. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI offers little enhancement to the simple act of physically handing over a form during a delivery route. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | This task requires in-person interaction with customers at physical locations (post offices, mail routes) and judgment about which forms are appropriate. Current AI has no capability to physically deliver cards or interact face-to-face with customers in real-world settings. |
| Task automatability | claude-sonnet-5 | 4/5 | Distributing standardized forms and directing customers to change-of-address processes is a simple, repetitive information task easily replicated by self-service kiosks, apps, or automated systems. |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | This task has strong inherent barriers: it requires human-customer contact (regulatory and practical), physical presence in postal facilities and on mail routes, and direct interaction that customers expect from USPS staff. Legal and service requirements mandate human involvement. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement exists for handing out forms; the main friction is customer habit and lack of digital access for some populations, not regulation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | There is no AI system that performs this task at any cost, making cost comparison infeasible. The human labor cost for form distribution remains the only practical option. |
| Cost vs. human wage | claude-sonnet-5 | 4/5 | Digital self-service forms cost far less than having a carrier physically distribute paper forms, though some print/mail infrastructure costs remain. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed product performs this task; it fundamentally requires human presence and physical form distribution at customer touchpoints, which is outside the scope of existing AI systems. |
| Technical feasibility today | claude-sonnet-5 | 4/5 | USPS already offers online change-of-address services and automated form dispensers at post offices, demonstrating this is reliably deployed today, though physical carrier handoff still occurs in some cases. |
Return incorrectly addressed mail to senders.
30CI 25–35 · exposure 25 · augmentation 50 · importance 4.0/5 · click for rater detail
Return incorrectly addressed mail to senders.
30| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | USPS modernization is slow-moving; mail handling remains largely manual and physical, with limited public evidence of AI-driven automation in sorting or return operations despite decades of opportunity. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Postal services have adopted automated sorting technology but the physical carrier task itself remains a low-digitization, labor-intensive process with slow automation uptake industry-wide. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI-assisted address reading and database lookup could help mail carriers more quickly identify correct sender addresses or flag suspicious cases, moderately raising their efficiency without replacing their judgment on ambiguous mail. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI-powered address verification and OCR systems used in mail centers help flag incorrect addresses before or after carrier handling, providing moderate assistance to the overall process. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | Identifying incorrect addresses requires reading and parsing envelope text, matching against databases, and determining sender information—tasks where current OCR and address-matching systems are partly functional but error-prone on handwritten or degraded addresses. End-to-end automation with 50% time savings at equal quality is not reliably achievable today. |
| Task automatability | claude-sonnet-5 | 2/5 | Identifying incorrectly addressed mail and physically redirecting it requires physical handling and judgment about ambiguous addresses; AI can assist with OCR/address validation but cannot perform the physical sorting/return end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | The USPS is a government agency with strict procedural and liability requirements around mail handling; returning mail to the wrong sender carries legal and service-quality consequences, creating strong organizational and regulatory barriers to unsupervised automation. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No strict licensing requirement, but physical infrastructure, logistics, and the need for a human presence on delivery routes create moderate practical barriers to full automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | OCR and address-lookup APIs are inexpensive per unit, but integrating them into mail-sorting infrastructure and managing exceptions requires significant setup and oversight; the labor cost of a mail carrier is already low for this repetitive task, limiting economic advantage. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | While automated sorting equipment is cost-effective at centralized facilities, replacing the carrier's physical handling and judgment on-route offers little cost advantage since a human must still be present to deliver mail. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | OCR systems and address validation tools exist in production (e.g., USPS address lookup), but they struggle with handwritten addresses, damaged mail, and edge cases; no deployed system fully automates the decision of whether to return mail without human review of ambiguous cases. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Automated mail sorting systems with OCR and address verification exist in postal processing centers, but the carrier-level task of identifying and physically returning misaddressed mail during delivery routes is not handled by deployed AI products. |
Bundle mail in preparation for delivery or transportation to relay boxes.
24CI 14–35 · exposure 20 · augmentation 25 · importance 4.3/5 · click for rater detail
Bundle mail in preparation for delivery or transportation to relay boxes.
24| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Postal services are traditionally slow to adopt automation in mail handling due to labor agreements, regulatory constraints, and the physical unpredictability of mail; adoption remains limited to narrow sorting stages, not bundling operations. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Postal delivery is a physical, low-digitization sector with slow automation adoption for granular field tasks like this, despite some backend automation in mail processing centers. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | AI could assist with address reading and sorting logic display to guide bundling, but the core task is physical manipulation where meaningful augmentation is limited without robotics; most assistance would be peripheral to the main operation. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI could optimize route planning or bundle sequencing but offers little direct assistance to the physical act of bundling mail by hand. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | Bundling mail requires physical manipulation of variable shapes and sizes, some reading of addresses and sorting data, and adaptive handling—capabilities that current AI and robotics cannot perform reliably end-to-end in the unstructured postal environment. While sorting logic can be automated, the physical bundling task itself remains predominantly manual. |
| Task automatability | claude-sonnet-5 | 2/5 | Physical bundling of mixed mail requires manipulation, sorting judgment, and handling of irregular items that current robotics/AI cannot reliably replicate outside controlled facility environments.rst. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Postal service operations are union-represented and heavily regulated; mail handling is explicitly a Postal Service employee responsibility under federal statute. Strong organizational and legal barriers protect this role from substitution. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No strict licensing barrier for the bundling itself, but it's embedded in a unionized government workflow with physical infrastructure constraints that slow any automation change. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | The capital cost of robotics capable of handling variable mail sizes, weights, and orientations, plus integration and maintenance, would far exceed the loaded wage of a postal worker performing this repetitive but semi-skilled task. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | Deploying robotic sorting/bundling systems at this granular, distributed scale would require significant capital investment exceeding the low hourly cost of a human carrier performing this quick task. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed commercial system reliably performs full mail bundling (physical sorting, grouping, wrapping, and staging) as a standalone task in production postal environments. Narrow sorting machines exist but do not replace the bundling operation itself. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Some automated sortation exists in mail processing centers, but bundling for individual carrier routes and relay boxes at the last-mile stage is still done manually by carriers today. |
Hold mail for customers who are away from delivery locations.
18CI 0–35 · exposure 13 · augmentation 25 · importance 4.4/5 · click for rater detail
Hold mail for customers who are away from delivery locations.
18| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | The task is deeply embedded in regulated government postal infrastructure with minimal digitization pressure; adoption of automation is not occurring in practice. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Postal services are traditionally slow adopters of AI for physical operations, though they have digitized request intake systems; the physical mail handling sector lags behind digital-native industries. |
| Augmentation potential | claude-haiku-4-5-20251001 | 1/5 | AI cannot meaningfully assist a postal worker in deciding whether to hold mail or managing physical custody of customer mailpieces. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI-driven scheduling and notification systems (like USPS Hold Mail online requests) already help carriers and clerks manage hold requests more efficiently, improving accuracy and reducing manual tracking errors. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | Holding mail for absent customers requires physical custody of items and discretionary judgment about customer intent—neither can be automated by current AI systems without full infrastructure redesign and human supervision. |
| Task automatability | claude-sonnet-5 | 2/5 | This task involves physical sorting, tracking, and storage of mail based on customer requests, which requires physical handling that current AI cannot perform end-to-end; only administrative scheduling aspects could be assisted. |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Federal postal services are heavily regulated, and mail custody is a legally protected responsibility with strict regulatory and liability requirements that mandate human accountability. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement exists for this task, but organizational logistics and physical infrastructure (mail rooms, hold bins) create moderate friction against full automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | A human postal worker performs this task as part of standard mail delivery; any AI system would require physical infrastructure, legal oversight, and human validation, costing more than the existing human-integrated approach. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | The administrative request-logging portion could be cheaply automated, but the physical sorting and holding of mail still requires human labor, keeping overall costs comparable to current staffing. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed postal automation system performs this task; it fundamentally requires human judgment, physical handling, and legal accountability for custodial decisions. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | While postal services use software to log hold requests, the actual identification, retrieval, and physical holding of mail items remains a manual process performed by carriers and facility staff. |
Leave notices telling patrons where to collect mail that could not be delivered.
17CI 10–24 · exposure 8 · augmentation 25 · importance 4.4/5 · click for rater detail
Leave notices telling patrons where to collect mail that could not be delivered.
17| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Postal services are traditional, geographically distributed, and heavily dependent on physical logistics. Adoption of autonomous systems for last-mile delivery tasks remains nascent and limited to narrow pilots in controlled environments. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | placeholder |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | AI could assist by automatically generating notice templates or optimizing which addresses receive notices based on delivery patterns, but the core task of physically traveling and leaving notices offers limited augmentation value while a human remains in the loop. |
| Augmentation potential | claude-sonnet-5 | 2/5 | placeholder |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | This task requires physical presence at customer locations, judgment about where to leave notices, and contextual decision-making about delivery failures. Current AI systems cannot autonomously navigate to addresses, physically leave notices, or make on-site decisions about placement. |
| Task automatability | claude-sonnet-5 | 2/5 | The physical act of leaving a paper notice at a specific address requires physical presence and manipulation, which current AI cannot perform; only the notice-generation portion is automatable."},"feasibility":{"rating":1,"rationale":"No deployed product performs the physical delivery of notices; this remains a manual field task requiring a human carrier."},"cost_ratio":{"rating":1,"rationale":"AI cannot perform the physical component at all, so there is no viable AI cost comparison for the full task; a human must still travel and place the notice."},"barriers":{"rating":2,"rationale":"No licensing barrier exists, but physical presence and USPS operational/regulatory processes create practical friction against automation."},"adoption_velocity":{"rating":1,"rationale":"Postal delivery is a low-digitization, physical-labor sector with minimal AI-driven displacement in last-mile delivery tasks."},"augmentation":{"rating":2,"rationale":"AI could help pre-print notices or route optimize the carrier's stops, but offers minimal help for the physical notice-leaving act itself."}}, |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | Postal service operations are regulated and unionized, creating organizational friction, though there are no hard legal barriers preventing automation research. Customer preference for human contact and liability concerns about autonomous systems accessing properties provide moderate friction. |
| Adoption barriers | claude-sonnet-5 | 2/5 | placeholder |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | The cost of an autonomous robot capable of navigating to customer addresses and physically placing notices, plus infrastructure integration, vastly exceeds the loaded wage of a mail carrier performing this task as part of their route. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | placeholder |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed product performs this end-to-end task today. While route planning and notice generation exist, the physical execution—traveling to addresses and physically leaving notices—remains purely human or robot-dependent, with no production systems doing this at scale. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | placeholder |
Sell stamps and money orders.
16CI 5–28 · exposure 13 · augmentation 25 · importance 3.8/5 · click for rater detail
Sell stamps and money orders.
16| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | The postal service and mail delivery remain low-digitization, traditionally human-centric sectors with slow adoption of autonomous retail automation. Most mail carriers continue direct sales with minimal AI-driven substitution. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Postal delivery is a low-digitization, physical-labor sector with minimal AI/robotic adoption for mobile point-of-sale functions. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | AI could offer minor assistance through inventory tracking or transaction logging tools, but the core task of selling stamps and money orders—requiring face-to-face customer interaction and payment handling—does not lend itself to meaningful AI augmentation of carrier productivity. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI could help with backend inventory tracking or mobile payment apps, but offers little direct enhancement to the physical act of selling stamps on a route. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | Selling stamps and money orders requires in-person transaction handling, payment processing, and customer interaction at postal counters or vehicles. Current AI systems cannot autonomously conduct these physical retail transactions or interact with customers face-to-face, making meaningful end-to-end automation infeasible today. |
| Task automatability | claude-sonnet-5 | 2/5 | This involves physical cash/card handling and in-person interaction while on a delivery route, which current AI systems cannot perform end-to-end; only the transaction-processing backend could be automated.' |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Selling stamps and money orders involves handling government-issued financial instruments and cash transactions, which carry liability, regulatory oversight of postal services, and implicit human-contact expectations from customers seeking official postal services. |
| Adoption barriers | claude-sonnet-5 | 3/5 | Money order sales involve financial handling and postal regulations, plus physical presence and trust in cash transactions, creating moderate procedural and organizational barriers to automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | Automating this task would require expensive in-person or remote service infrastructure (kiosks, agents, payment systems) that would exceed the modest loaded wage of mail carrier transaction time, making full automation economically unfavorable. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | Physical automation (a machine or robot carrier) would require significant capital investment far exceeding the marginal wage cost of a human already on the route performing this small task. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed product reliably automates the sale of stamps and money orders by mail carriers in production environments. While stamp vending machines exist, they are separate infrastructure, not AI systems performing this task, and mail carriers' direct sales remain human-driven. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Kiosks and vending machines for stamps exist but a mail carrier selling stamps/money orders curbside is a physical, mobile transaction with no deployed robotic or AI product replicating it. |
Obtain signed receipts for registered, certified, and insured mail, collect associated charges, and complete any necessary paperwork.
14CI 5–23 · exposure 13 · augmentation 38 · importance 4.7/5 · click for rater detail
Obtain signed receipts for registered, certified, and insured mail, collect associated charges, and complete any necessary paperwork.
14| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Postal service is a heavily regulated, traditional sector with slow digitization and organizational inertia. Adoption of autonomous mail-handling solutions is minimal; most carriers still use paper-based or basic mobile workflows in production. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Postal delivery is a highly physical, low-digitization sector where AI/robotic adoption for last-mile delivery and in-person transactions remains at pilot stages at best. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI could meaningfully assist by auto-populating paperwork, calculating charges, and flagging compliance issues, raising carrier productivity on the administrative portions. However, the signature and charge-collection elements require human judgment and customer interaction, limiting augmentation scope. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI could assist with digital paperwork, route optimization, or scanning/logging receipts, but it offers minimal help with the core physical task of signature collection and payment handling. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI could generate paperwork and manage charge calculations, the core requirement—obtaining signed receipts from customers—demands physical presence and real-time human interaction that current systems cannot fully automate. This task fundamentally hinges on in-person verification and signature capture, which remains a bottleneck. |
| Task automatability | claude-sonnet-5 | 1/5 | This requires physical presence to hand-deliver mail, obtain a physical or electronic signature from a specific recipient, and collect payment in person—none of which current AI systems can perform end-to-end without a robotic/physical embodiment. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Significant barriers exist: regulatory requirements for verified signatures on registered/certified/insured mail, legal liability for lost or mishandled items, and postal service regulations that mandate authorized personnel collect receipts and charges. Customer preference for human verification of valuable shipments adds friction. |
| Adoption barriers | claude-sonnet-5 | 4/5 | Postal regulations, chain-of-custody requirements for registered/insured mail, and financial handling (collecting charges) impose significant procedural and legal barriers requiring an authorized human carrier. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Even with automated paperwork and charge calculation, the requirement for in-person signature collection means a human (either the carrier or a customer service agent) must remain in the loop, keeping total cost comparable to or higher than current human-only performance. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no AI system that can substitute for the physical delivery and collection labor involved, so AI cost is not comparable—human labor remains the only viable option today. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | No deployed AI product reliably handles the full end-to-end workflow of customer interaction, signature verification, charge collection, and paperwork completion in production mail delivery contexts. Digital signature apps exist but don't encompass the full task scope or integrate seamlessly into postal workflows at scale. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product autonomously performs physical mail delivery, in-person signature collection, and cash/charge handling; this remains firmly in the human physical labor domain. |
Report any unusual circumstances concerning mail delivery, including the condition of street letter boxes.
12CI 5–19 · exposure 8 · augmentation 25 · importance 3.3/5 · click for rater detail
Report any unusual circumstances concerning mail delivery, including the condition of street letter boxes.
12| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Postal services are traditional, government-operated monopolies with slow digital transformation relative to private tech firms. Adoption of autonomous inspection systems is minimal and pilots are not widespread in the sector. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Postal delivery is a low-digitization, physical-labor sector with minimal AI agent deployment for field observation tasks. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | Mobile apps or image-logging tools could assist mail carriers in documenting observations, but current AI adds limited value beyond what a human inspector seeing the scene in person can already determine and report directly. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI could help carriers log or transcribe reports via voice-to-text or mobile apps, offering minor assistance but not transforming the core observational task. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | Detecting unusual circumstances and assessing the physical condition of mail boxes requires on-site visual inspection, contextual judgment, and real-world navigation that current AI cannot perform autonomously. While image analysis might assist in identifying obvious damage, the task fundamentally requires a human present at the location making contextual judgments about what qualifies as 'unusual.' |
| Task automatability | claude-sonnet-5 | 2/5 | Reporting requires physical observation during a delivery route, which AI cannot yet perform end-to-end; only the reporting/documentation portion (e.g., dictating a note) is automatable. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Postal services are heavily regulated government agencies with strict accountability and liability requirements for mail handling and infrastructure reporting. The human mail carrier's physical presence and signature/report carries legal and operational weight that automated systems cannot easily replace without regulatory change. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No strict licensing barrier for the reporting itself, but the task is inherently tied to physical presence on a route, creating a practical (not regulatory) barrier to automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | AI systems cannot yet perform this task end-to-end, making any cost comparison premature. The on-site presence and judgment required means human mail carriers remain the necessary baseline. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | AI has no physical means to inspect street conditions, so it cannot substitute at any cost; a human carrier must be present regardless. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed product reliably performs autonomous mail-delivery anomaly detection and street-box condition assessment in production. Computer vision systems exist for controlled image analysis, but they cannot substitute for the situated human judgment required during daily mail routes. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product autonomously observes and reports mailbox/street conditions during a physical delivery route; this remains a human observational task. |
Deliver mail to residences and business establishments along specified routes by walking or driving, using a combination of satchels, carts, cars, and small trucks.
7CI 5–10 · exposure 0 · augmentation 25 · importance 4.6/5 · click for rater detail
Deliver mail to residences and business establishments along specified routes by walking or driving, using a combination of satchels, carts, cars, and small trucks.
7| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Postal services, especially USPS, are slow-moving organizations with strong institutional and union constraints. Adoption of autonomous delivery remains minimal despite years of pilot projects; no meaningful displacement of mail carriers has occurred. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Postal/logistics last-mile delivery is a physically-intensive, low-digitization sector where full automation adoption is minimal and mostly experimental. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | Route-optimization software and GPS assist human mail carriers in planning, but AI adds limited productivity gain to the core physical delivery task itself. Route planning assistance is modest compared to the dominance of walking/driving and door-to-door placement. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI can assist with route optimization and sorting logistics, but offers little direct augmentation to the physical act of walking/driving and delivering mail itself. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | Mail delivery requires navigating complex physical environments, handling diverse mail formats, and interacting with gates, doors, and barriers that vary unpredictably—capabilities far beyond current AI. No autonomous system reliably performs door-to-door delivery at scale today. |
| Task automatability | claude-sonnet-5 | 1/5 | Physical delivery of mail to specific addresses via walking or driving requires embodied manipulation, navigation of varied terrain/entryways, and physical handoff that current AI systems cannot perform end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Significant barriers exist: USPS is a regulated government entity with statutory labor protections, union agreements govern mail-carrier roles, and liability for lost/damaged mail creates accountability asymmetry favoring human delivery. Customer preference for human contact and regulatory oversight of the mail system itself add friction. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No strict licensing requires a human specifically, but physical/regulatory barriers exist (right-of-way access, mailbox regulations, safety rules for autonomous vehicles, liability for lost/misdelivered mail) creating real friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | Autonomous vehicles and robots capable of reliable last-mile delivery are expensive to acquire, maintain, and operate. Current loaded human wage for mail carriers is substantially cheaper than the all-in cost of autonomous systems. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | Autonomous delivery hardware (robots, drones) plus required infrastructure and oversight currently costs far more per delivery than a human carrier's wage for this task. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed autonomous mail-delivery product performs this task reliably in production. Autonomous delivery remains largely experimental; human mail carriers remain the standard across postal services. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product autonomously walks routes and delivers mail into mailboxes at scale; delivery drones/robots remain pilot-stage and narrow in scope (e.g., flat packages, controlled environments). |
Return to the post office with mail collected from homes, businesses, and public mailboxes.
5CI 0–10 · exposure 0 · augmentation 25 · importance 4.7/5 · click for rater detail
Return to the post office with mail collected from homes, businesses, and public mailboxes.
5| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | The postal service is a traditional, heavily regulated sector with strong union representation and slow technology adoption. No measurable displacement of mail carriers by automation is occurring in production today. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Postal delivery and collection is a physical, low-digitization sector with minimal AI-driven displacement; automation efforts (e.g., delivery robots) remain pilot-stage and rare in production. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | GPS routing and vehicle tracking systems marginally assist with route optimization, but AI does not meaningfully augment the core physical collection and transport task itself. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI could assist with route optimization or scheduling, but it offers little direct assistance to the physical act of collecting and returning mail to the post office. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | This task requires physically collecting mail from geographically distributed locations and returning it to a specific facility. Current AI systems have no capability to perform physical collection and transport without human operation of a vehicle. |
| Task automatability | claude-sonnet-5 | 1/5 | This requires physical driving/walking a route and transporting physical mail back to a facility, which current AI systems cannot perform; it requires embodied physical action, not information processing. |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | USPS mail collection and transport is a federally regulated function; only authorized postal carriers may collect mail from public and residential mailboxes. Legal authorization and liability requirements create hard barriers to automation. |
| Adoption barriers | claude-sonnet-5 | 3/5 | While not legally restricted to licensed humans, there are logistical, safety, security (mail chain-of-custody), and union/regulatory considerations that create moderate friction against pure automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | The capital cost of autonomous vehicles equipped for mail collection, plus ongoing maintenance and operational overhead, far exceeds the cost of employing a human mail carrier for this task. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no AI system that performs this physical transport task, so AI cost is not comparable; any automation would require costly autonomous vehicle/robotics infrastructure far exceeding human labor costs today. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed commercial product performs this end-to-end task today. While autonomous vehicles exist in limited settings, none reliably handle the full workflow of mail collection from multiple outdoor locations and return transport. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed AI product autonomously collects and transports physical mail; this remains a manual physical logistics task performed by human carriers or vehicles. |
Turn in money and receipts collected along mail routes.
3CI 0–5 · exposure 0 · augmentation 25 · importance 4.4/5 · click for rater detail
Turn in money and receipts collected along mail routes.
3| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Mail carrier operations remain low-digitization, field-based work with no observable automation of cash-handling tasks in production postal services. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Postal delivery is a physical, low-digitization sector with minimal AI/robotic adoption for physical cash handling and route-based collection tasks. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | Digital tracking and reconciliation software could assist with receipt logging and balance verification, but the physical cash handling and delivery steps remain human-required. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI could help with tracking, reconciling, or logging collected amounts digitally, but offers little assistance with the core physical collection and transfer task. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | This task requires physical handling of currency and receipts, manual cash reconciliation, and secure delivery to postal facilities. Current AI systems cannot perform the physical collection, transport, or verification steps end-to-end. |
| Task automatability | claude-sonnet-5 | 1/5 | This requires physically collecting cash/receipts along a route and delivering them to a facility, a physical handling and transport task that current AI cannot perform end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Financial institutions and postal regulations impose strict liability and accountability requirements for handling currency; a licensed, bonded human must legally receive and account for cash. |
| Adoption barriers | claude-sonnet-5 | 4/5 | Handling of money and financial accountability typically requires trusted, bonded, authorized personnel with chain-of-custody and audit responsibilities, creating strong organizational and trust-based barriers. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | Automating cash handling would require hardware (robotics, secure transport), infrastructure, and oversight systems that far exceed the cost of a mail carrier performing this task manually. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no AI substitute for the physical act of collecting and transporting money, so AI cost comparison is not applicable and effectively more expensive/infeasible. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed AI system reliably performs the full task of collecting, reconciling, and turning in cash and receipts from field routes. This remains entirely human-dependent in production. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed AI product performs physical collection and turn-in of money/receipts; this remains entirely a human physical task. |
Sign for cash-on-delivery and registered mail before leaving the post office.
0CI 0–0 · exposure 0 · augmentation 0 · importance 4.5/5 · click for rater detail
Sign for cash-on-delivery and registered mail before leaving the post office.
0| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | This task is embedded in postal regulatory compliance; it is not subject to sector-wide AI adoption patterns because the barrier is legal/mandatory human sign-off, not economic substitution. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Postal service physical operations are a low-digitization, high physical-presence sector with minimal AI/agent adoption for custody and signature tasks. |
| Augmentation potential | claude-haiku-4-5-20251001 | 1/5 | AI offers no meaningful assistance to signing for mail. The task is already minimal and procedural; there is no analytical or drafting component where AI could enhance productivity. |
| Augmentation potential | claude-sonnet-5 | 1/5 | AI offers no meaningful assistance to the physical act of signing for and taking custody of mail before departure. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | Signing for mail is a legal/procedural requirement that establishes accountability and chain of custody. This requires a human signature and explicit acknowledgment that cannot be meaningfully automated without losing the accountability mechanism itself. |
| Task automatability | claude-sonnet-5 | 1/5 | This requires a physical human presence to accept custody and legally sign for high-value/tracked mail; no AI system can perform this physical, accountability-bearing act.' |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Postal regulations and chain-of-custody laws require a human employee to personally sign for and assume responsibility for registered and COD mail; this is a hard legal requirement that prevents automation regardless of technical capability. |
| Adoption barriers | claude-sonnet-5 | 5/5 | This involves legal chain-of-custody and accountability for registered/COD mail, requiring an authorized human employee to sign and accept liability, a hard institutional and legal barrier. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | A brief human signature and acknowledgment is near-zero cost. Any AI system attempting to handle the liability or verification would be more expensive than the seconds a human requires. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no AI substitute performing this physical signing/custody task, so cost comparison favors the human by default since AI cannot execute it at all. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed AI system can generate legally valid signatures or assume liability for registered/COD mail accountability. This task requires human assumption of responsibility, which no AI product performs in production. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product performs physical custody transfer and signature accountability for mail carriers; this remains entirely a human physical/administrative act. |
Related occupations — Office & Administrative Support
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.