Passenger Attendants
53-6061.00Median wage $37,720/yr27,110 employed (US)Rank #485 of 923 scored · top 53% by substitution
Provide services to ensure the safety of passengers aboard ships, buses, trains, or within the station or terminal. Perform duties such as explaining the use of safety equipment, serving meals or beverages, or answering questions related to travel.
Sub-scores
0–100 · band = confidence interval from rater disagreement
Substitution28
Exposure25
Augmentation36
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
12 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
17%
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%25
panel mean rating 2.0/5 → substitution pressure 25/100
Technical feasibility todayw 20%25
panel mean rating 2.0/5 → substitution pressure 25/100
Cost vs. human wagew 15%28
panel mean rating 2.1/5 → substitution pressure 28/100
Adoption barriersw 20%inverted — strong barriers lower the score35
panel mean rating 3.6/5 (barrier strength) → substitution pressure 35/100
Sector adoption velocityw 10%24
panel mean rating 2.0/5 → substitution pressure 24/100
Task breakdown (12 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.
Provide customers with information on routes, gates, prices, timetables, terminals, or concourses.
77CI 75–79 · exposure 75 · augmentation 63 · importance 4.7/5 · click for rater detail
Provide customers with information on routes, gates, prices, timetables, terminals, or concourses.
77| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 4/5 | Airlines, transit agencies, and airports have rapidly deployed chatbots, mobile apps, and kiosk systems over the past 3–5 years. Information services are among the fastest-adopted automation tasks in transportation; production displacement is already measurable. |
| Sector adoption velocity | claude-sonnet-5 | 4/5 | Transportation and travel sectors have rapidly adopted digital self-service tools, apps, and AI chatbots for passenger information, though full replacement of human attendants remains partial. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can assist human attendants by providing instant access to unified route/gate/pricing databases, reducing lookup time and supporting handling of complex multi-leg queries. However, the task is fundamentally information retrieval, so augmentation impact is limited compared to tasks requiring judgment. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI-powered apps and assistants significantly augment human attendants by handling routine queries, freeing them to focus on complex or in-person passenger needs. |
| Task automatability | claude-haiku-4-5-20251001 | 4/5 | Current AI systems (chatbots, voice assistants, knowledge retrieval agents) can reliably retrieve and deliver route, gate, price, and timetable information end-to-end with minimal human intervention. Integration with airport/transit systems enables automated answers that meet the 50% time-saving threshold; however, occasional edge cases (system outages, special circumstances) may require human escalation. |
| Task automatability | claude-sonnet-5 | 4/5 | Providing route, gate, price, and schedule information is a lookup/communication task well-suited to AI chatbots and kiosks integrated with transit databases, meeting the time-saving threshold for most routine inquiries. |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | No licensing or legal requirement mandates a human provide this information; no liability asymmetry prevents automation. Minor barriers exist: customer preference for human contact in some contexts and organizational inertia in legacy operations, but these do not legally prevent substitution. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing is required to give route/schedule information, though some contexts (e.g., accessibility assistance, security-sensitive areas) still favor human presence for trust and liability reasons. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | AI inference cost for information retrieval is negligible (fractions of cents per query), while human attendant labor costs $15–25/hour loaded. At-scale automation achieves orders-of-magnitude cost reduction per task instance. |
| Cost vs. human wage | claude-sonnet-5 | 4/5 | Automated information systems (apps, kiosks, chatbots) cost far less per interaction than staffing human attendants for routine informational queries. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 4/5 | Deployed products (airline chatbots, airport information systems, voice assistants) already perform this task at scale in production environments with high reliability. Mature systems handle the bulk of routine inquiries; error rates are low for factual lookups, though some legacy systems have integration gaps. |
| Technical feasibility today | claude-sonnet-5 | 4/5 | Airlines, airports, and transit systems already deploy chatbots, apps, and digital signage/kiosks that reliably provide this information in production at scale, though live human attendants still handle edge cases and disruptions. |
Count and verify tickets and seat reservations and record numbers of passengers boarding and disembarking.
72CI 70–75 · exposure 70 · augmentation 63 · importance 4.1/5 · click for rater detail
Count and verify tickets and seat reservations and record numbers of passengers boarding and disembarking.
72| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 4/5 | Major airlines (United, Delta, Southwest, international carriers) have deployed automated boarding and counting systems across significant portions of their fleets over the past decade; this is mainstream adoption in commercial aviation, not pilot-stage. |
| Sector adoption velocity | claude-sonnet-5 | 4/5 | Transportation sectors (airlines, transit, rail) have rapidly adopted automated ticketing, e-gates, and passenger-counting sensors as standard infrastructure over the past decade. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | Automated systems assist attendants by providing real-time passenger counts and flagging discrepancies for manual resolution, raising accuracy and reducing manual counting labor; the human remains in the loop for exception handling and final reconciliation. |
| Augmentation potential | claude-sonnet-5 | 4/5 | Automated systems can feed real-time verified counts and reservation data to attendants, reducing manual tallying and letting them focus on passenger service and safety tasks. |
| Task automatability | claude-haiku-4-5-20251001 | 4/5 | Counting passengers boarding/disembarking can be largely automated through computer vision systems at gates or cabin sensors; ticket/reservation verification is routine data-matching against airline systems. Most of the task meets the ≥50% time-saving bar, though physical verification of tickets for non-digital or damaged documents may require human oversight. |
| Task automatability | claude-sonnet-5 | 4/5 | Ticket verification and passenger counting are structured data-matching tasks well-suited to automated scanning and counting systems (e.g., barcode/RFID scanners, computer vision headcounts) that can operate with high accuracy and speed.: |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | No regulatory requirement mandates human ticket verification; airlines can legally automate boarding and counting. Minimal legal barriers exist, though some airlines maintain manual verification for customer experience or legacy operational reasons, creating only weak organizational friction. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No strict licensing requirement for this specific counting/verification function, though safety-related duties for passenger attendants (e.g., emergency procedures) may impose broader role requirements that limit full attendant replacement. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 4/5 | Automated gate scanners and counting sensors cost thousands per gate plus integration, but amortized across thousands of daily flights, the per-flight cost is a small fraction of a passenger attendant's hourly wage; oversight and exception-handling overhead is minimal. |
| Cost vs. human wage | claude-sonnet-5 | 4/5 | Automated scanning/counting hardware and software have low marginal per-passenger cost compared to a human attendant's wage, though upfront hardware and integration costs exist. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 3/5 | Airlines deploy gate counting systems and boarding pass scanners in production, but end-to-end automation faces gaps: manual reconciliation of count discrepancies, handling of edge cases (standby passengers, gate agents overriding), and integration with legacy reservation systems remains partially manual and error-prone at scale. |
| Technical feasibility today | claude-sonnet-5 | 4/5 | Automated ticket scanners, e-gates, and passenger-counting sensors are already deployed at scale in airlines, trains, and buses, reliably verifying tickets and counting boarding/disembarking passengers. |
Determine or facilitate seating arrangements.
44CI 32–55 · exposure 38 · augmentation 75 · importance 3.9/5 · click for rater detail
Determine or facilitate seating arrangements.
44| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 3/5 | Airlines have invested in seating systems for booking, but gate-level facilitation by attendants remains largely manual. Some adoption of assignment tools exists, but widespread replacement in production workflows remains limited. |
| Sector adoption velocity | claude-sonnet-5 | 3/5 | Transportation/hospitality sectors have adopted automated booking and seat-assignment systems widely, but the attendant's in-person facilitation role has seen slower AI-driven change. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI seat-mapping and optimization tools can significantly assist flight attendants by suggesting efficient arrangements, flagging accessibility needs, and identifying conflicts, allowing attendants to execute decisions faster while maintaining oversight of special accommodations. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI-driven seating algorithms and apps significantly help attendants pre-plan and optimize arrangements, letting them focus on exceptions and personalized service. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | Seating arrangements require dynamic judgment about passenger needs, safety regulations, and aircraft configuration. While basic seat selection can be automated, resolving conflicts, accommodating special needs, and handling real-time changes demand human flexibility that current AI cannot reliably manage end-to-end. |
| Task automatability | claude-sonnet-5 | 3/5 | Seating arrangement logic (matching preferences, group needs, mobility constraints) can be automated via reservation/booking algorithms, but real-time human adjustments and conflict resolution on-site still require a person present. |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | Passenger safety and satisfaction create moderate organizational friction, and airline operational procedures favor human attendants for conflict resolution and accommodation. No hard legal requirement exists, but customer preference and liability concerns slow automation. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement specifically for seating decisions, though customer service expectations and accessibility/safety accommodations create some organizational friction against full automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | AI seating optimization software requires significant integration with airline systems and human oversight for exceptions. The total cost of deployment, maintenance, and required human intervention exceeds the wage cost of a flight attendant performing this task. |
| Cost vs. human wage | claude-sonnet-5 | 3/5 | Software-based seat assignment is cheap at scale, but the physical presence and interpersonal facilitation aspect of the task still requires paid staff, keeping blended cost comparable rather than dramatically cheaper. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Seat assignment systems exist in airline software, but the full task of determining and facilitating arrangements—handling accessibility needs, family grouping, rebooking disruptions, and passenger requests—remains predominantly human-performed with limited AI deployment in production. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | Airlines and transit systems already use automated seat-assignment software extensively, but on-the-spot facilitation (disputes, special needs, last-minute changes) still relies on human attendants. |
Greet passengers boarding transportation equipment and announce routes and stops.
37CI 25–50 · exposure 30 · augmentation 50 · importance 4.1/5 · click for rater detail
Greet passengers boarding transportation equipment and announce routes and stops.
37| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Despite technological possibility, the transport sector has adopted automation slowly for passenger-facing roles. Most airlines, transit agencies, and coaches retain human attendants; adoption remains in pilot phases for automated announcements with minimal displacement of staff. |
| Sector adoption velocity | claude-sonnet-5 | 3/5 | Transit and airline sectors have adopted automated announcement systems fairly widely, but full replacement of attendant roles remains slow due to safety and accessibility mandates. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI-assisted route announcements and passenger information systems can support attendants, reducing repetitive speech and freeing time for other duties, but the core greeting and interaction tasks remain human-centric and difficult to meaningfully augment with current systems. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI-driven announcement and translation systems can support attendants by handling routine route/stop announcements, letting them focus on passenger service and safety. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI systems can generate route announcements via text-to-speech, the personalized greeting and real-time social interaction component of boarding passengers requires human presence and responsiveness. Only narrow parts of the task (pre-recorded announcements) can be automated without significant quality loss or safety concerns. |
| Task automatability | claude-sonnet-5 | 2/5 | Automated announcements already exist widely, but greeting passengers and handling live variability (delays, questions, safety needs) still benefits from human presence; only the announcement sub-part is fully automatable.4 |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Transit regulations typically mandate human staff for passenger safety, security checks, and assistance; liability and customer-facing service requirements create strong organizational and regulatory barriers to full automation. Many jurisdictions legally require human attendants on passenger transport. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement blocks automated announcements, but some transportation contexts require staff presence for safety/assistance (e.g., disability accommodations), creating moderate organizational friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Text-to-speech and simple announcement systems are inexpensive, but they cannot replace the full attendant role (greeting, customer interaction, safety oversight). The marginal cost savings from automating announcements alone do not justify equipment and integration costs relative to attendant wages. |
| Cost vs. human wage | claude-sonnet-5 | 4/5 | Automated announcement systems are cheap to run per trip compared to a dedicated attendant, though the human greeting/assistance function still requires paid staff. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Deployed systems can produce route announcements via automation, but no production system reliably performs the full interpersonal greeting and boarding management task. The human-contact and safety-critical aspects remain operator-dependent in all major transit systems. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | Automated PA/route-announcement systems are deployed at scale on buses, trains, and planes, but the greeting/personal-interaction component is not replaced by any deployed product. |
Respond to passengers' questions, requests, or complaints.
35CI 25–45 · exposure 30 · augmentation 63 · importance 4.2/5 · click for rater detail
Respond to passengers' questions, requests, or complaints.
35| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 3/5 | Airlines have begun deploying chatbots and automated response systems, but adoption remains mixed and mostly for pre-flight and simple queries. Full replacement of attendant-passenger interaction is not yet widespread despite the digitization of the sector. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Transportation and hospitality sectors show slower AI adoption for front-line, in-person passenger service roles compared to purely digital customer service channels. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI tools can effectively assist flight attendants by providing real-time answers to FAQs, policy summaries, and complaint resolution frameworks, allowing them to focus on empathetic engagement and complex problem-solving. This augmentation noticeably raises attendant productivity without removing the human from the interaction. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI tools like translation apps, quick-reference systems, and chat-based support can help attendants answer questions faster and access information, meaningfully aiding but not replacing their role. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI can handle routine, templated passenger inquiries (e.g., flight delays, seat assignments), responding to complex complaints, personal emergencies, or nuanced requests requires empathy, judgment, and real-time context that current systems struggle with. Most passenger interactions require human presence and cannot achieve 50% time savings at equal quality today. |
| Task automatability | claude-sonnet-5 | 2/5 | AI chatbots can handle basic scripted queries, but resolving in-person complaints, reading emotional cues, and improvising solutions on transportation vehicles require physical presence and human judgment that current AI cannot fully replicate. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Airlines face regulatory oversight (FAA, safety protocols), passenger expectation of human contact for complaints, and liability concerns around poor service recovery or misinformation. Customer preference for human empathy in difficult situations creates organizational friction against full automation. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No strict licensing requirement for this specific interaction, but safety regulations often mandate attendants be present, and customer service quality/liability concerns create moderate friction against full automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 4/5 | AI-driven chatbots have low per-interaction inference costs compared to human attendants' loaded wages (salary + benefits), though integration and monitoring overhead reduces the ratio. At scale, AI is substantially cheaper per interaction handled. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | AI could cheaply handle simple FAQ-type queries, but the need for a physically present human attendant for most interactions means overall cost savings are limited since staffing can't be eliminated. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 3/5 | Chatbots and conversational AI products exist in airline settings and can resolve simple requests, but they frequently mishandle ambiguous or emotionally charged complaints and often defer to human agents. Production systems handle a narrow subset of interactions with material error and escalation rates. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | While airlines and transit companies deploy chatbots for pre-trip questions, no product reliably handles real-time, in-person passenger complaints or requests during travel with acceptable quality. |
Explain and demonstrate safety procedures and safety equipment use.
17CI 9–25 · exposure 20 · augmentation 50 · importance 4.1/5 · click for rater detail
Explain and demonstrate safety procedures and safety equipment use.
17| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | The heavily regulated aviation industry has shown minimal substitution of AI or automation for live safety briefings; regulatory mandates and safety liability concerns ensure humans remain responsible for this task across all carriers. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Transportation/passenger service sectors are relatively slow adopters of AI for physical, safety-critical, human-contact tasks compared to information-sector benchmarks. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI could assist by generating pre-flight briefing scripts, creating supplementary visual aids, or providing real-time prompting on passenger questions, but the attendant remains the primary delivery mechanism and decision-maker for safety communication. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI-generated video content, multilingual translation, and digital signage can support and standardize safety communication, but the human attendant remains central to executing and enforcing the task. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI could generate safety scripts or generate instructional videos, the task fundamentally requires real-time demonstration of physical equipment (oxygen masks, life vests, emergency exits) and responsive interaction with passengers. Current AI cannot reliably perform hands-on demonstration or adapt to diverse passenger questions and comprehension levels in real time, falling well short of 50% time savings at equal quality. |
| Task automatability | claude-sonnet-5 | 2/5 | While pre-recorded safety briefings and video demonstrations already exist, the live, adaptive, and physically embodied aspects (checking compliance, assisting passengers, handling exceptions) still require human presence and cannot be fully automated end-to-end today. |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Aviation regulations (FAA, EASA, ICAO standards) explicitly mandate that a qualified flight attendant must personally conduct safety demonstrations and be available to respond to passenger questions. This legal and safety requirement creates a hard barrier to automation. |
| Adoption barriers | claude-sonnet-5 | 4/5 | Safety regulations (e.g., FAA, transit authorities) often mandate a live attendant or crew member responsible for safety compliance and liability, creating strong regulatory and liability barriers to full automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | The cost of developing, deploying, and maintaining an AI system (including hardware for demonstrations, cameras, sensors, and continuous oversight) to perform live safety demonstrations would exceed the wage cost of a single flight attendant conducting the briefing. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | Pre-recorded video content is cheap, but the human attendant is still needed for compliance monitoring, physical demonstration, and passenger interaction, so total cost savings from AI substitution are limited. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed product reliably replaces a human attendant demonstrating and explaining safety procedures to a cabin full of passengers. AI chatbots or pre-recorded videos cannot substitute for the regulatory requirement of a trained human performing live safety briefings and equipment demonstrations. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Automated safety videos are deployed in aviation, but interactive demonstration, physical guidance, and real-time passenger assistance remain done by humans in production settings; no robotic/AI agent handles this reliably at scale. |
Open and close doors for passengers.
16CI 0–32 · exposure 17 · augmentation 13 · importance 4.0/5 · click for rater detail
Open and close doors for passengers.
16| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Aviation is a heavily regulated, conservative industry with slow adoption cycles; this specific task remains firmly in human-operator territory with no meaningful industry shift toward automation. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Transportation service roles involving physical passenger assistance are low-digitization, low-AI-adoption environments with minimal automation of physical tasks like this. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | While sensor assistance or alerts about door status could provide minor support, the task itself is straightforward manual operation that offers limited scope for AI assistance to meaningfully increase attendant productivity. |
| Augmentation potential | claude-sonnet-5 | 1/5 | AI offers essentially no productivity assistance for the physical act of opening and closing a door for passengers. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While automatic door systems exist, passenger attendants typically operate cabin doors in aircraft that require human judgment about timing, safety checks, and passenger flow management. Current AI cannot reliably perform the full task including safety assessment and coordination with other crew. |
| Task automatability | claude-sonnet-5 | 1/5 | This is a physical manipulation task requiring bodily presence at a door with passengers boarding/alighting; no off-the-shelf AI system performs this physical action end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Aviation safety regulations strictly govern cabin door operation, and federal law requires human crew oversight of passenger safety procedures, creating hard legal and liability barriers to full automation of this safety-critical function. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No licensing requirement specifically for door-opening, but safety regulations, liability concerns, and the need for human judgment in emergencies or assisting disabled passengers create meaningful friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Retrofitting aircraft doors with fully autonomous, fail-safe systems that meet aviation safety standards would be extremely expensive compared to a flight attendant's loaded wage, making the cost-benefit unfavorable. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | Replacing this with robotics or automation would require costly physical infrastructure investment far exceeding the marginal wage cost of a human attendant performing this simple action. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 3/5 | Automatic door systems are deployed in many settings (retail, airports), but passenger attendants' door operations in aviation cabins specifically remain primarily manual due to safety protocols and regulatory requirements that haven't shifted to full automation. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product opens/closes physical doors for passengers as a substitute for human attendants; automated doors exist as engineered systems but not as an 'AI performing an attendant's task' product. |
Perform equipment safety checks prior to departure.
11CI 0–23 · exposure 13 · augmentation 25 · importance 4.5/5 · click for rater detail
Perform equipment safety checks prior to departure.
11| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Passenger aircraft operators are highly regulated, risk-averse, and move slowly on safety protocol changes. No meaningful adoption of AI for pre-departure safety checks is evident in the commercial aviation sector, which prioritizes human sign-off and legal accountability. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Transportation/passenger service sectors performing physical safety checks show minimal AI adoption for this specific hands-on task, as it remains physical and safety-regulated with low digitization in execution. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | AI could potentially assist by highlighting anomalies on cameras for crew review, but current systems are not sufficiently accurate to augment safety-critical inspection in any meaningful way, and aviation culture prioritizes human judgment over AI suggestions for safety. |
| Augmentation potential | claude-sonnet-5 | 2/5 | Sensors and checklists software can support tracking and documentation, but the core physical verification act sees limited AI-driven productivity enhancement. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | Current AI systems cannot reliably perform end-to-end safety checks requiring visual inspection, physical verification, and judgment about equipment condition. While computer vision can detect some anomalies, safety-critical checks demand 100% accuracy, and existing systems fall far short of the 50% time-saving threshold without substantial human oversight. |
| Task automatability | claude-sonnet-5 | 1/5 | This requires physical inspection of equipment (seatbelts, doors, safety gear) in a real-world environment, which current AI systems cannot perform end-to-end without robotic embodiment.atibility remains far from equal-quality automation. |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Aviation regulation (FAA/EASA) legally mandates that a qualified human crew member sign off on safety checks before departure; this is non-delegable liability and a hard regulatory requirement, creating an absolute barrier to autonomous automation. |
| Adoption barriers | claude-sonnet-5 | 4/5 | Safety-critical inspections are often governed by regulatory and liability requirements mandating trained human verification, creating strong barriers to full automation even if sensor-based monitoring assists. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | The cost of AI vision systems, installation, integration, and required oversight by trained personnel would substantially exceed the cost of a flight attendant performing manual checks, especially given the need for redundancy and regulatory compliance in safety-critical contexts. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no viable AI substitute performing this physical task, so any hypothetical AI solution (robotics, sensors) would require costly hardware exceeding current human labor costs for this narrow function. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | No deployed product performs pre-departure safety checks autonomously in production at airlines. Existing computer vision and robotics projects exist at research or pilot stage, but they lack the reliability, regulatory approval, and integration necessary for real-world deployment on passenger aircraft. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product performs physical passenger safety equipment checks autonomously in production; this remains a manual, hands-on task performed by trained staff. |
Adjust window shades or seat cushions at the request of passengers.
7CI 0–15 · exposure 0 · augmentation 0 · importance 3.8/5 · click for rater detail
Adjust window shades or seat cushions at the request of passengers.
7| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Aviation is a highly regulated, safety-critical sector with minimal autonomous equipment adoption for passenger-facing tasks; the physical and regulatory constraints make this a laggard sector for automation. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Passenger service roles in transportation are low-digitization, physically embodied jobs with minimal AI/robotic adoption for such micro-tasks. |
| Augmentation potential | claude-haiku-4-5-20251001 | 1/5 | There is no meaningful way for AI to assist a human flight attendant in adjusting window shades or seat cushions, as these are simple, direct physical tasks that require minimal cognitive input. |
| Augmentation potential | claude-sonnet-5 | 1/5 | AI offers no meaningful assistance for this specific physical, on-demand adjustment task. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | Adjusting physical window shades or seat cushions requires mobile manipulation in a dynamic, confined space with irregular passenger requests; current AI systems have no reliable embodied robotics deployed on aircraft for this task. |
| Task automatability | claude-sonnet-5 | 1/5 | This is a physical manipulation task requiring presence in a vehicle/vessel and responsiveness to verbal requests; no current AI system can physically adjust shades or cushions. |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Strict FAA and international aviation safety regulations require human flight attendants to be present and authorized to perform cabin service; autonomous systems would face certification barriers and passenger safety concerns that legally prohibit unattended operation. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement exists, but the task requires physical presence and interpersonal service, creating practical (not regulatory) barriers to automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | Developing and deploying a robotic system capable of safe, reliable physical manipulation in aircraft cabins would cost orders of magnitude more than the hourly wage of a flight attendant performing these simple adjustments. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no AI substitute performing this physical task, so AI cost is not comparable—human labor is the only viable option today. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No commercial product or deployed system performs this task autonomously on aircraft today; the combination of physical manipulation, safety constraints, and real-time passenger interaction remains outside current practical deployment. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product performs physical passenger service adjustments; this remains firmly in the human physical labor domain. |
Signal transportation operators to stop or to proceed.
7CI 0–14 · exposure 8 · augmentation 25 · importance 3.9/5 · click for rater detail
Signal transportation operators to stop or to proceed.
7| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Transportation operators (airlines, buses, trains) continue to employ human attendants for safety signaling with minimal automation; adoption of autonomous systems for this critical safety function remains virtually absent in practice. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Transportation attendant roles involving physical safety signaling are in a low-digitization, physically-oriented sector with minimal AI agent adoption for this specific function. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | AI could assist with status monitoring or alerts to attendants, but the core task of human-to-operator signaling offers limited augmentation potential since it requires direct human judgment and real-time communication. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI could provide sensor-based alerts or camera monitoring to assist attendants in detecting hazards, but it does not meaningfully transform the core signaling task itself. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While simple signal generation is technically feasible, the task involves real-time safety-critical decisions that require immediate response to dynamic conditions, passenger positioning, and operator attention—factors that current AI systems cannot reliably monitor and respond to in real time without human oversight. |
| Task automatability | claude-sonnet-5 | 1/5 | This is a real-time physical safety signaling action requiring physical presence and immediate judgment on a moving vehicle, which current AI cannot perform end-to-end.itude |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Strong legal and safety barriers exist: transportation safety regulations require a human attendant to be responsible for passenger safety and to communicate with operators; automation would face strict regulatory requirements and liability asymmetry if a system fails. |
| Adoption barriers | claude-sonnet-5 | 4/5 | Safety-critical signaling in transit typically involves regulatory requirements, liability concerns, and reliance on human presence/judgment near passengers and vehicles, creating strong barriers to automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | The infrastructure required to automate passenger attendant signals (sensors, communication systems, safety validation) would exceed the cost of a human attendant performing the task. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no viable AI substitute performing this physical task, so cost comparison favors the human by default since no AI alternative exists in production. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed product reliably performs autonomous signaling to transportation operators in production safety-critical environments; this remains a human responsibility due to liability and safety requirements. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product has passenger attendants replaced by AI for live stop/proceed signaling to vehicle operators; this remains a human physical safety function. |
Secure passengers for transportation by buckling seatbelts or fastening wheelchairs with tie-down straps.
3CI 0–5 · exposure 0 · augmentation 0 · importance 4.7/5 · click for rater detail
Secure passengers for transportation by buckling seatbelts or fastening wheelchairs with tie-down straps.
3| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Aviation is a heavily regulated, safety-critical sector with strong regulatory barriers to automation of passenger safety tasks. Adoption of autonomous systems for physical restraint of passengers is negligible. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Passenger transport/attendant services are a physically-oriented, low-digitization sector with essentially no AI/robotic adoption for physical passenger securing tasks. |
| Augmentation potential | claude-haiku-4-5-20251001 | 1/5 | AI offers no meaningful assistance for physically securing passengers; the task is inherently manual and hands-on, with no intermediate steps where AI tools could enhance human productivity. |
| Augmentation potential | claude-sonnet-5 | 1/5 | AI offers no meaningful assistance for the physical act of buckling seatbelts or fastening tie-down straps. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | Securing passengers—buckling seatbelts or fastening wheelchairs with physical tie-downs—requires dexterous physical manipulation in varied, real-world conditions that current robots cannot reliably perform. This task cannot be automated end-to-end by available AI systems today. |
| Task automatability | claude-sonnet-5 | 1/5 | This requires physical manipulation of seatbelts and wheelchair tie-down straps on real passengers, which is a hands-on physical task no current AI system can perform. |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | This task has hard legal and safety barriers: federal aviation regulations require human flight attendants to conduct safety procedures, and liability for failed passenger restraint in an emergency is severe. A licensed human must legally perform or directly oversee this task. |
| Adoption barriers | claude-sonnet-5 | 4/5 | Passenger safety, liability for improper securing of vulnerable passengers (e.g., wheelchair users), and physical presence requirements create strong barriers against any non-human substitution. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | Robotic systems capable of safe, reliable physical manipulation of passengers would require substantial hardware investment and integration costs far exceeding the loaded wage of a flight attendant performing this task. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no AI-based substitute performing this physical task, so AI cost is not comparable; a human must be paid to do this regardless. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed commercial product performs this physical safety-critical task autonomously. The combination of variability (different body types, wheelchair configurations, mobility challenges) and safety liability makes this research-stage only. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product performs physical securing of passengers; this remains entirely a human physical-labor task with no robotic automation in production for this specific service context. |
Provide boarding assistance to elderly, sick, or injured people.
3CI 0–5 · exposure 0 · augmentation 13 · importance 4.4/5 · click for rater detail
Provide boarding assistance to elderly, sick, or injured people.
3| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Adoption velocity in this context is near-zero; airlines have not deployed automated systems for passenger boarding assistance. The physical, safety-critical, and human-dignity aspects of the task, combined with regulatory environment, mean production deployment remains extremely limited. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Passenger service and transit assistance roles are physical, low-digitization jobs with minimal AI agent deployment or measured displacement to date. |
| Augmentation potential | claude-haiku-4-5-20251001 | 1/5 | Augmentation potential is minimal because the task's core value—physical, compassionate support to vulnerable people—does not decompose into components where AI can meaningfully assist a human attendant. Informational pre-screening might help slightly, but does not transform the attendant's productivity on the boarding assistance itself. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI can help with scheduling, communication, or identifying passengers needing assistance in advance, but offers little direct help during the physical act of boarding assistance itself. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | Providing physical boarding assistance to vulnerable passengers requires human presence, judgment about individual mobility needs, and physical contact that current AI systems cannot perform. The task is fundamentally embodied and cannot be meaningfully automated without replacing the human attendant entirely—which is not feasible with current technology. |
| Task automatability | claude-sonnet-5 | 1/5 | This task requires physical presence, manual assistance (e.g., wheelchair transfer, guiding, lifting), and situational judgment for vulnerable people, none of which current AI systems can perform end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Strong regulatory, safety, and liability barriers exist: airlines have legal duty of care obligations, customers expect and often require human assistance, and automated systems assisting vulnerable passengers would face intense scrutiny and potential certification requirements. Human contact is effectively mandated by industry practice and customer need. |
| Adoption barriers | claude-sonnet-5 | 4/5 | Safety regulations, liability for injury during physical assistance, and ADA/accessibility mandates in transit require trained human staff to perform this hands-on task, creating strong barriers to automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | A physical attendant performing this task costs far less than developing, deploying, and maintaining a robotic system capable of safely assisting elderly or injured passengers, plus the liability and oversight costs associated with autonomous physical assistance. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no AI substitute performing this physical task, so AI cost is not comparable—human labor remains the only viable option, making AI effectively more expensive (infinite) for full task replacement. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed AI product can physically assist passengers with boarding today. This task requires embodied interaction, physical support, and real-time adaptation to individual conditions that exceed the capabilities of current robotic or autonomous systems in production airline environments. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product provides physical boarding assistance to elderly, sick, or injured passengers; this remains purely a human physical-service task. |
Related occupations — Transportation & Material Moving
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.