Flight Attendants
53-2031.00Median wage $63,580/yr131,650 employed (US)Rank #771 of 923 scored · top 84% by substitution
Monitor safety of the aircraft cabin. Provide services to airline passengers, explain safety information, serve food and beverages, and respond to emergency incidents.
Sub-scores
0–100 · band = confidence interval from rater disagreement
Substitution16
Exposure16
Augmentation30
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
25 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
0%
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%17
panel mean rating 1.7/5 → substitution pressure 17/100
Technical feasibility todayw 20%14
panel mean rating 1.6/5 → substitution pressure 14/100
Cost vs. human wagew 15%16
panel mean rating 1.6/5 → substitution pressure 16/100
Adoption barriersw 20%inverted — strong barriers lower the score18
panel mean rating 4.3/5 (barrier strength) → substitution pressure 18/100
Sector adoption velocityw 10%9
panel mean rating 1.4/5 → substitution pressure 9/100
Task breakdown (25 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.
Inspect passenger tickets to verify information and to obtain destination information.
62CI 30–95 · exposure 67 · augmentation 38 · importance 3.4/5 · click for rater detail
Inspect passenger tickets to verify information and to obtain destination information.
62| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Airlines are highly regulated and conservative with passenger-facing operations; while digital ticket scanning has grown, human attendants remain the norm for final boarding verification. Industry adoption of full automated ticket inspection remains minimal, with most airlines treating this as a human task despite digitization elsewhere. |
| Sector adoption velocity | claude-sonnet-5 | 5/5 | Airlines have already deeply adopted automated boarding and ticket verification systems industry-wide for years, representing mature, fast adoption. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI-powered ticket scanning and data extraction could assist flight attendants by automatically reading and flagging anomalies, reducing manual note-taking and cross-referencing time. However, the human would still make the final verification decision, so augmentation is useful but not transformative for this narrowly scoped task. |
| Augmentation potential | claude-sonnet-5 | 2/5 | Since this sub-task is largely already automated via kiosks/scanners rather than performed manually by attendants, there is limited additional augmentation value for the human task itself. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI could read and verify ticket information from images or digital formats, the task requires human-like interaction with passengers, real-time decision-making about anomalies, and integration with airline systems that lack standard APIs. Current systems cannot reliably handle this end-to-end with 50% time savings at equal quality. |
| Task automatability | claude-sonnet-5 | 5/5 | Ticket verification is a simple structured-data lookup task already handled by boarding scanners, e-ticket systems, and check-in kiosks with essentially full automation. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Regulatory requirements mandate that airlines maintain verified passenger manifests and boarding processes; liability for security and safety breaches creates strong error-cost asymmetry. Customer expectations and safety regulations typically require human verification of identity and boarding eligibility, creating both legal and organizational friction against full automation. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement mandates a human specifically verify tickets; the main friction is that flight attendants perform this incidentally alongside safety duties that do require certification. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | AI systems require significant infrastructure investment (cameras, processing, database integration, and ongoing maintenance), while ticket inspection is a brief, repetitive task performed by flight attendants who are already salaried for other duties. The cost per task-equivalent remains higher than the marginal human labor involved. |
| Cost vs. human wage | claude-sonnet-5 | 5/5 | Automated scanning hardware/software costs pennies per passenger versus staff time, making it dramatically cheaper than manual ticket checking. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 3/5 | Optical character recognition and document classification systems exist and work reasonably well on tickets, but they struggle with damaged, non-standard, or international documents, and lack robust integration with ticketing databases for real-time verification. Deployed products can handle routine cases but fail on edge cases common in live operations. |
| Technical feasibility today | claude-sonnet-5 | 5/5 | Airlines universally deploy automated boarding pass scanners and check-in systems that verify passenger identity and destination reliably at scale in production today. |
Prepare reports showing places of departure and destination, passenger ticket numbers, meal and beverage inventories, the conditions of cabin equipment, and any problems encountered by passengers.
60CI 55–65 · exposure 66 · augmentation 75 · importance 4.0/5 · click for rater detail
Prepare reports showing places of departure and destination, passenger ticket numbers, meal and beverage inventories, the conditions of cabin equipment, and any problems encountered by passengers.
60| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Airlines are information-intensive and digitally advanced, but flight operations remain heavily regulated and crew-dependent. Adoption of automated reporting has been slow despite high digitization; most carriers still rely on manual crew reporting and mobile apps rather than autonomous systems, reflecting institutional conservatism and safety culture. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Airlines are a highly regulated, moderately digitized but operationally conservative sector; back-office reporting automation is adopted unevenly and slower than in software/finance sectors. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI can significantly assist flight attendants by auto-populating reports from airline systems, flagging inventory issues, and organizing passenger complaints into structured forms. This reduces clerical burden and helps attendants focus on safety-critical observations while maintaining their essential role in validation and judgment. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI-assisted apps and mobile tools can pre-fill inventories, ticket data, and cabin condition checklists, significantly speeding up report preparation while the attendant still adds incident-specific details. |
| Task automatability | claude-haiku-4-5-20251001 | 4/5 | Flight attendants manually observe and record data about departure/destination, tickets, inventory, equipment, and passenger issues. Most of this information can be automatically extracted from digital systems (airline databases, inventory tracking), flight data, and incident reports—or scraped from cabin sensors and cameras. A structured reporting workflow would enable near-complete automation with minimal human input, easily exceeding the 50% time-saving threshold. |
| Task automatability | claude-sonnet-5 | 4/5 | This is largely structured data compilation (ticket numbers, inventories, cabin conditions) that current AI/automation could handle via forms, scanning, and templated report generation, though the 'problems encountered' narrative portion requires human summarization judgment. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Aviation safety and regulatory requirements (FAA, international air transport regulations) mandate formal reporting and often require crew sign-off on safety-critical observations. Airlines are also liable for incident documentation accuracy, creating legal and safety guardrails that typically require human validation or authorization before automation fully replaces the task. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement mandates a human write this report, but airline internal policy, safety documentation standards, and accountability for incident reporting create some organizational friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 4/5 | Automating data collection and report generation costs only inference and integration overhead, whereas flight attendant labor is substantial. Once deployed, the cost per report would be a small fraction of the hourly wage saved, easily achieving favorable cost ratios. |
| Cost vs. human wage | claude-sonnet-5 | 4/5 | Automated data logging and report templating is far cheaper than manual compilation by flight attendant labor time, though some human input/oversight is still needed for accuracy and incident narratives. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 3/5 | Products exist to extract structured data from airline systems and digital logbooks, and some carriers use mobile apps for incident reporting. However, capturing passenger complaints and equipment observations reliably still requires human verification in most deployed systems; end-to-end automation remains inconsistent across the aviation industry at scale. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | Airlines already use digital systems and apps for post-flight reporting with some automated data capture, but full end-to-end AI generation of these mixed structured/narrative reports isn't a standard deployed product specifically for this task. |
Answer passengers' questions about flights, aircraft, weather, travel routes and services, arrival times, or schedules.
52CI 25–79 · exposure 50 · augmentation 63 · importance 3.6/5 · click for rater detail
Answer passengers' questions about flights, aircraft, weather, travel routes and services, arrival times, or schedules.
52| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 4/5 | The travel and airline industry is highly digitized and early-adopting; major carriers have deployed chatbots, mobile apps, and IVR systems for years, and adoption of AI-powered passenger assistance tools is accelerating across the industry. Production deployment is common among large airlines and growing among mid-tier carriers. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Airlines have adopted chatbots and apps for pre-flight customer service but in-flight passenger interaction remains largely unautomated, reflecting slow adoption in this specific task context. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI systems augment flight attendants by handling high-volume routine queries and freeing them to focus on complex passenger needs, safety, and hospitality. Voice or text agents can assist attendants in real-time with instant access to flight data, weather, and service options, substantially raising their efficiency and responsiveness. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI-powered apps, translation tools, and knowledge bases can help attendants quickly retrieve flight status, weather, or routing info to answer passengers faster and more accurately. |
| Task automatability | claude-haiku-4-5-20251001 | 4/5 | Current AI systems can reliably answer factual questions about flights, aircraft specifications, weather conditions, routes, services, and schedules by querying real-time flight databases and knowledge bases. While some edge cases and complex passenger situations may require human judgment, the vast majority of straightforward information requests can be automated with >50% time savings, especially with chatbots or voice agents integrated into airline systems. |
| Task automatability | claude-sonnet-5 | 2/5 | While AI chatbots and apps can answer generic flight/schedule questions, in-flight or in-person passenger interactions require real-time physical presence, situational awareness, and trust-building that current AI cannot replicate end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | Airlines face minimal regulatory or legal barriers to automating factual information delivery; customer preference for human interaction and some safety-critical context (emergency procedures, medical situations) create modest friction, but there is no licensing requirement or legal mandate for a human to answer these routine questions. |
| Adoption barriers | claude-sonnet-5 | 4/5 | FAA regulations mandate minimum numbers of certified flight attendants onboard for safety and passenger service, creating a hard regulatory barrier to full automation of this in-cabin role. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | AI inference cost for answering routine passenger questions is negligible (pennies per interaction including oversight), while flight attendant labor costs $20–35/hour loaded. AI systems achieve at least 100× cost advantage per task completion, especially at scale across high-volume airline operations. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | Digital self-service tools are cheap for basic queries, but replacing the human presence requirement in-cabin still requires a paid attendant, so overall cost savings for this specific task are limited. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 4/5 | Deployed airline chatbots and virtual assistants already handle this task in production at scale (e.g., major carriers use AI for FAQ responses, booking confirmations, and schedule queries). These systems reliably answer routine questions; occasional failures occur with ambiguous queries or system integration gaps, but the technology is mature and proven in real-world deployment. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Airlines deploy chatbots for pre-flight FAQs and app-based schedule info, but no product substitutes for a flight attendant answering questions in-cabin during flight operations. |
Announce flight delays and descent preparations.
43CI 25–60 · exposure 45 · augmentation 63 · importance 4.0/5 · click for rater detail
Announce flight delays and descent preparations.
43| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Airlines have been slow to automate announcements despite technical feasibility; regulatory caution, safety culture, and passenger expectations for human contact keep adoption limited to niche automation pilots. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Airlines have been slow to adopt full automation of onboard crew communication tasks, though some pre-recorded and app-based passenger notifications are increasingly used for delays. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI can draft, schedule, and deliver announcements to assist flight attendants in coordinating timing and consistency across cabin crews, freeing them to focus on passenger service and safety oversight. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI-driven scheduling and flight-tracking systems can feed real-time delay and descent information to attendants, helping them deliver more accurate and timely announcements. |
| Task automatability | claude-haiku-4-5-20251001 | 4/5 | AI could generate and deliver announcements about flight delays and descent preparations with high consistency and 50%+ time savings, though current systems lack the real-time integration with flight management systems and regulatory compliance oversight that would push this to full automation today. |
| Task automatability | claude-sonnet-5 | 2/5 | Announcements can be scripted and even generated/read by automated systems, but this is a small, situational part of a broader in-person safety and service role that also requires real-time judgment and physical presence.' |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | FAA and international aviation regulations require trained flight crew to maintain command of passenger communications and safety announcements; airlines face liability and passenger-trust concerns if automation replaces the human presence entirely. |
| Adoption barriers | claude-sonnet-5 | 3/5 | Aviation safety regulations require certified flight attendants onboard, and passenger communication during emergencies or delays is tied to broader safety and regulatory duties, creating moderate structural barriers. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 4/5 | AI inference and integration costs are minimal compared to the loaded wage of a flight attendant performing announcement duties, making AI substantially cheaper for this isolated task. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | The announcement itself could be cheaply automated, but since it's a minor task within a role requiring a human present for many other duties, there's no standalone cost savings from automating just this piece. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 3/5 | Text-to-speech and chatbot systems can produce announcements reliably in controlled conditions, but deployed airline systems still rely heavily on human flight attendants for nuance, passenger context, and regulatory adherence; no airline has fully automated this at scale. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Automated PA systems and pre-recorded announcements exist on aircraft, but live, context-specific delay/descent updates in production are still delivered by crew due to variability and passenger interaction needs. |
Operate audio and video systems.
37CI 14–60 · exposure 41 · augmentation 50 · importance 3.7/5 · click for rater detail
Operate audio and video systems.
37| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Airlines have been slow to automate cabin-crew tasks due to regulatory conservatism, union agreements, and passenger expectations for human service, despite technical feasibility; early automation remains in niche areas (entertainment systems) rather than widespread operator displacement. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Airline cabin crew operations are a physically-embedded, highly regulated, low-digitization environment with minimal AI adoption for hands-on equipment operation. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI can significantly assist flight attendants by automating routine audio/video operations (announcements, system status checks, entertainment scheduling), freeing them to focus on safety and passenger service while maintaining human oversight and emergency response. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI could help generate or standardize scripted announcements or troubleshoot AV issues via guidance, but offers little transformative assistance to the physical operation task itself. |
| Task automatability | claude-haiku-4-5-20251001 | 4/5 | Operating audio and video systems on aircraft is highly procedural and standardized, with clear on/off sequences, volume adjustments, and preset configurations. Current AI systems can handle pre-flight checks, in-flight announcements, and video playback with minimal human oversight, achieving >50% time savings once integrated into aircraft systems. |
| Task automatability | claude-sonnet-5 | 2/5 | Physically operating cabin audio/video equipment (switches, PA systems) requires in-person manipulation of hardware, which AI cannot perform without robotics; only minor sub-elements like automated announcements could be scripted. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Commercial aviation is heavily regulated by the FAA and EASA; any automation of safety-critical systems (emergency announcements, communication systems) requires certification, liability frameworks, and regulatory sign-off, creating substantial barriers to deployment independent of human operators. |
| Adoption barriers | claude-sonnet-5 | 3/5 | While not licensed specifically for this sub-task, flight attendants are required by aviation safety regulations to be present and operationally trained, creating organizational and regulatory friction against automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 4/5 | Once integrated into existing avionics systems, AI-driven audio/video operation costs are marginal (compute on aircraft systems already running), making the per-task cost orders of magnitude lower than paying a flight attendant's loaded wage for this single task. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no AI substitute performing this physical task, so cost comparison favors the human by default since no viable AI alternative exists. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 3/5 | Automated audio/video control systems exist in modern aircraft (e.g., inflight entertainment automation), but they typically require human fallback for troubleshooting, system failures, and passenger requests, and are not yet fully autonomous across all scenarios in production. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed AI product operates physical in-cabin AV equipment; this remains a manual, hands-on task performed by crew. |
Attend preflight briefings concerning weather, altitudes, routes, emergency procedures, crew coordination, lengths of flights, food and beverage services offered, and numbers of passengers.
36CI 0–73 · exposure 45 · augmentation 63 · importance 4.3/5 · click for rater detail
Attend preflight briefings concerning weather, altitudes, routes, emergency procedures, crew coordination, lengths of flights, food and beverage services offered, and numbers of passengers.
36| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Airlines operate in a highly regulated, conservative sector with strong safety and union protections. Adoption of AI to replace human briefing attendance is slow and limited to assisting (not replacing) crew due to regulatory compliance and operational culture. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Airlines, while adopting AI in scheduling and customer service, have not applied any automation to physical crew briefing attendance, a highly regulated, human-centric process. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI can meaningfully assist flight attendants by pre-generating summaries, highlighting critical updates, organizing briefing materials, and creating personalized checklists, substantially raising efficiency while the crew member remains the responsible party for briefing comprehension and acknowledgment. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI can help summarize weather, route data, or generate briefing materials in advance, aiding preparation, but the attendance and live coordination itself is not augmented in real time. |
| Task automatability | claude-haiku-4-5-20251001 | 5/5 | Attending and comprehending preflight briefings is primarily information reception and retention. Current AI can reliably extract, summarize, and organize briefing information from documents, audio, or text; an AI system could generate standardized briefing summaries, checklists, and alerts that meet the ≥50% time-saving threshold by replacing manual note-taking and information synthesis. |
| Task automatability | claude-sonnet-5 | 1/5 | This is an in-person, synchronous team briefing requiring physical presence, live coordination with crew, and situational awareness that cannot be replaced by current AI systems. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Federal Aviation Regulations (FARs) and airline operating procedures mandate that all crew members attend and acknowledge understanding of safety briefings; regulatory and safety liability requirements create a legal requirement for human attestation and sign-off, significantly restricting full automation substitution. |
| Adoption barriers | claude-sonnet-5 | 5/5 | Aviation safety regulations require certified crew members to physically attend briefings and be accountable for safety procedures, making this a hard regulatory and licensing barrier. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | The inference and integration cost of an AI system to summarize briefings and generate checklists is negligible compared to the loaded wage of a flight attendant, making the cost ratio heavily favorable to automation. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no AI product performing this task, so no cost comparison favors AI; the human attendance is mandatory and irreplaceable. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 4/5 | Deployed products (speech-to-text, document summarization, AI assistants) can already parse and organize briefing data reliably in production settings. However, real-time integration into live crew workflows and validation against current operational standards remain partially manual, so it falls short of 5. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product substitutes for a human crew member attending and participating in a live preflight briefing; this remains entirely a human activity. |
Take inventory of headsets, alcoholic beverages, and money collected.
31CI 23–39 · exposure 33 · augmentation 38 · importance 3.1/5 · click for rater detail
Take inventory of headsets, alcoholic beverages, and money collected.
31| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Airlines have shown minimal adoption of automated inventory systems for in-flight galleys. The task remains manual across the industry, with no evidence of production-scale AI or sensor-based automation in commercial operations. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Airlines are digitizing onboard sales via tablets and POS systems but full automation of inventory reconciliation remains slow-moving within a highly regulated, safety-focused industry. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | Simple barcode or image recognition could assist in recording headset counts or beverage depletion, but the small scope of the task and the need for immediate, on-aircraft reconciliation limits the productivity uplift compared to a manual 10-minute count. |
| Augmentation potential | claude-sonnet-5 | 3/5 | Digital POS and inventory apps already help flight attendants track sales and counts more efficiently, reducing manual tallying even though a human still performs physical counts and cash handling. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | Inventory counting of physical items can be partially automated (using cameras or weight sensors for beverages, RFID for headsets), but the task involves securing cash and reconciling it, which requires human verification and accountability. Current AI cannot reliably handle the full end-to-end workflow with sufficient security and compliance. |
| Task automatability | claude-sonnet-5 | 3/5 | Counting and reconciling inventory is a simple, structured data-entry task that AI-enabled scanning/POS systems could handle, but it currently requires a human physically checking carts and cash onboard. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Cash handling and financial reconciliation are heavily regulated; airline policies typically require a named employee to be accountable for beverage and money inventory. Liability and regulatory requirements create strong barriers to full automation without human sign-off. |
| Adoption barriers | claude-sonnet-5 | 3/5 | Handling cash and alcohol involves compliance, security, and liability considerations that create some institutional friction, though no strict licensing requirement mandates a human for this specific inventory task. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Installing and maintaining camera, weight sensor, or RFID systems on aircraft is capital-intensive, and the recurring oversight cost approaches or exceeds the 10–15 minutes of human labor typically needed for this inventory task on a flight. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | Hardware (scanners, RFID tags, POS terminals) and integration costs for automating this narrow task are nontrivial relative to the few minutes a flight attendant spends on it as part of broader duties. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | While barcode/RFID scanning and image recognition systems exist, no deployed product reliably performs complete inventory reconciliation of heterogeneous items (headsets, beverages, cash) in an aircraft galley environment with the accuracy and security needed. Most systems require significant manual oversight. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Some airlines use digital point-of-sale systems that automatically track beverage/headset sales, but full automated physical inventory reconciliation with cash counting is not a mature deployed product on aircraft today. |
Collect money for meals and beverages.
28CI 25–30 · exposure 30 · augmentation 50 · importance 3.6/5 · click for rater detail
Collect money for meals and beverages.
28| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Airlines have adopted mobile payment and card readers to streamline the process, but full automation of payment collection is not occurring; attendants still perform the task manually at scale. Adoption is incremental optimization rather than displacement. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Airlines are a physical, safety-regulated industry with slow technology adoption cycles for cabin service processes; contactless payment is emerging but full automation of this task is not underway. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | Mobile POS systems and payment apps assist flight attendants by reducing manual cash handling, speeding up transactions, and providing inventory tracking. These tools meaningfully improve efficiency while keeping the attendant in the loop. |
| Augmentation potential | claude-sonnet-5 | 3/5 | Mobile POS devices and apps already help attendants process payments faster and track inventory, offering moderate productivity assistance while the attendant still performs the interaction. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While payment processing and item tracking could be partially automated via mobile systems or card readers, the interpersonal aspects of offering items, handling cash, and resolving payment issues in a confined aircraft environment require human presence. Current systems cannot replace the full end-to-end task of collecting payment in-flight with 50% time savings. |
| Task automatability | claude-sonnet-5 | 2/5 | Physical collection of payment on an aircraft requires physical presence and handling of a point-of-sale device or cash; AI can process payment transactions but cannot physically walk the aisle and interact with passengers in-cabin.' |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Federal Aviation Regulations require flight attendants to be present and responsible for passenger safety and service during flight. Customer expectation for human interaction during payment, combined with regulatory requirements for in-flight staff, creates substantial adoption barriers. |
| Adoption barriers | claude-sonnet-5 | 4/5 | FAA/regulatory safety requirements mandate a minimum number of certified flight attendants onboard regardless of automation of ancillary tasks like payment collection, creating a strong structural barrier. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Mobile payment systems reduce overhead but do not eliminate the need for a flight attendant to be present during the collection process. The cost of AI-powered systems plus human oversight remains comparable to or higher than the labor cost of attendants performing the task themselves. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | Self-service payment terminals could reduce some labor cost, but flight attendants remain required onboard for safety, so AI payment processing alone doesn't eliminate the human cost, making net savings modest. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 3/5 | Payment processing products (mobile POS, card readers) exist and are deployed in airlines, but they still require a human attendant to initiate transactions, handle cash, and manage exceptions. No product autonomously collects payments without human intermediation. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Automated payment kiosks and tap-to-pay systems exist in some contexts, but no deployed product autonomously conducts in-flight beverage/meal sales without a human attendant present. |
Inspect and clean cabins, checking for any problems and making sure that cabins are in order.
16CI 5–28 · exposure 13 · augmentation 25 · importance 3.8/5 · click for rater detail
Inspect and clean cabins, checking for any problems and making sure that cabins are in order.
16| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Aviation is a highly regulated, capital-constrained, and safety-conscious sector with strong union presence and strict operational procedures. Adoption of autonomous cabin inspection and cleaning systems remains in pilot stages with minimal production deployment. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Airlines are a highly regulated, safety-critical physical services sector with minimal AI-driven automation of cabin physical tasks; adoption of AI here is essentially nonexistent. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI-powered inspection tools (image recognition, checklists, route optimization) could assist flight attendants in identifying problems faster and ensuring comprehensive coverage, raising efficiency moderately while humans remain responsible for final sign-off and problem-solving. |
| Augmentation potential | claude-sonnet-5 | 1/5 | AI offers negligible assistance for the physical acts of inspecting and cleaning a cabin, though checklists could be digitized without meaningfully changing task performance. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI-powered robotic systems could eventually inspect cabins, current deployed automation cannot reliably perform both inspection and cleaning end-to-end with 50% time savings. Visual inspection via computer vision is feasible, but physical cleaning and problem diagnosis (identifying mechanical issues, safety hazards) across cabin complexity remains beyond practical automation today. |
| Task automatability | claude-sonnet-5 | 1/5 | This is a physical inspection and cleaning task requiring bodily presence, manual dexterity, and mobility inside an aircraft cabin, which current AI systems cannot perform. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Regulatory barriers are substantial: FAA and international aviation authorities require certified personnel to sign off on aircraft safety and cabin condition before flight. Airlines carry liability for cabin defects and passenger safety, creating high error costs that push toward human accountability and oversight. |
| Adoption barriers | claude-sonnet-5 | 3/5 | While no strict licensing requirement bars automation of cleaning itself, aviation safety regulations require crew to visually verify cabin safety conditions before flight, creating regulatory and safety-related friction against pure automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Robotic cleaning and inspection systems are capital-intensive and require significant integration and maintenance. Current costs per flight cycle would likely exceed the labor cost of a flight attendant performing the task, even accounting for multiple flights per attendant per day. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no AI system substituting for this physical labor, so cost comparison favors humans by default; any robotic solution would be far costlier than a flight attendant's marginal time on this task. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | No production systems currently perform full cabin inspection and cleaning autonomously. Prototype robotic cleaners and inspection drones exist in research/pilot phases, but they lack the dexterity, contextual judgment, and reliability needed for airline certification and liability requirements. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed AI product physically inspects or cleans airplane cabins; this remains firmly a human/robotic-hardware task with no mature automation solution in service. |
Determine special assistance needs of passengers, such as small children, the elderly, or persons with disabilities.
15CI 5–25 · exposure 13 · augmentation 38 · importance 4.3/5 · click for rater detail
Determine special assistance needs of passengers, such as small children, the elderly, or persons with disabilities.
15| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Airlines operate in a heavily regulated, safety-sensitive environment with strong organizational and legal pressure to retain human decision-making on passenger care. Adoption of AI for this safety-critical task remains negligible despite digitization in other areas. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Airlines are adopting AI for booking, scheduling, and customer service chatbots, but in-cabin physical safety tasks show minimal AI adoption. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI could assist flight attendants by highlighting passengers who appear to need help (via computer vision alerts) or offering decision-support information about accommodations, thereby raising efficiency. However, augmentation is limited by the human-essential nature of the interaction itself. |
| Augmentation potential | claude-sonnet-5 | 2/5 | Pre-flight manifests or apps could flag passengers with known needs, offering minor informational assistance, but the actual determination remains a human, contextual task. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | AI could assist with some detection tasks (e.g., identifying mobility aids, age appearance) but determining nuanced assistance needs requires real-time interaction, empathy, and judgment about individual circumstances. Current computer vision and LLMs lack reliable context awareness and cannot engage passengers in conversation to assess actual needs, making end-to-end automation infeasible. |
| Task automatability | claude-sonnet-5 | 1/5 | Assessing special assistance needs requires real-time in-person observation, judgment, and interpersonal interaction with passengers, which current AI cannot perform physically on an aircraft. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Federal aviation regulations (FAA, TSA) assign explicit duty-of-care responsibilities to flight attendants for passenger safety and accommodation; crew must be present and accountable. Liability and regulatory requirements make substitution of AI for human judgment on passenger needs a hard barrier. |
| Adoption barriers | claude-sonnet-5 | 4/5 | Safety regulations (FAA/EASA) require trained cabin crew to identify and manage passengers needing special assistance, and liability for safety failures is high. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Implementing computer vision, analysis infrastructure, and required human oversight to validate AI determinations would likely cost as much or more than having flight attendants perform the task directly, given the safety-critical nature and low volume per interaction. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no AI substitute performing this physical, in-person task, so cost comparison favors the human by default. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Vision systems can detect some indicators (wheelchairs, canes, children) but deployed products do not reliably assess the full spectrum of special needs or integrate with airline systems in production. No mature end-to-end product exists for this task; any application would require substantial human oversight and validation. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed AI product performs in-cabin passenger needs assessment; this remains entirely a human physical and social task. |
Check to ensure that food, beverages, blankets, reading material, emergency equipment, and other supplies are aboard and are in adequate supply.
11CI 0–23 · exposure 13 · augmentation 25 · importance 4.2/5 · click for rater detail
Check to ensure that food, beverages, blankets, reading material, emergency equipment, and other supplies are aboard and are in adequate supply.
11| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Airlines have shown minimal adoption of automated supply verification systems, relying instead on human-led checklists. The highly regulated nature of aviation and low digitization incentive for this specific task mean adoption velocity remains very low. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Airlines are a highly regulated, safety-critical physical-world sector with minimal AI adoption for hands-on cabin safety inspections. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | Limited augmentation potential exists: mobile apps or sensors could flag low inventory levels to crew members, but the core task—physical verification and judgment—remains primarily human-centered. Current tools offer minimal productivity gains over traditional visual inspection checklists. |
| Augmentation potential | claude-sonnet-5 | 2/5 | Digital checklists or inventory tracking apps could mildly assist in logging supply status, but the core physical inspection is not meaningfully augmented by current AI. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI could theoretically assist with inventory tracking via computer vision or RFID sensors, the task requires physical verification of items in varied spatial configurations within an aircraft cabin, which current deployed systems cannot reliably do end-to-end. The task demands tactile confirmation and judgment about supply adequacy that remains difficult to automate fully. |
| Task automatability | claude-sonnet-5 | 1/5 | This requires physical walkthrough of an aircraft cabin to visually inspect and count physical supplies and safety equipment, which current AI cannot perform without robotic embodiment.' |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Regulatory barriers are substantial: FAA regulations and airline safety protocols typically mandate human crew sign-off on pre-flight supply checks for liability and safety reasons. Automation would require regulatory approval and certification, and airlines remain cautious about removing human oversight from safety-critical tasks. |
| Adoption barriers | claude-sonnet-5 | 5/5 | FAA and other aviation regulators mandate certified crew members perform pre-flight safety and supply checks, making this a hard licensing and regulatory barrier. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Current AI systems capable of assisting with inventory checks (cameras, sensors, integration) would be expensive to deploy across a fleet relative to the labor cost of a single pre-flight checklist performed by an existing crew member. The infrastructure cost exceeds the modest wage savings from partial automation. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no viable AI substitute performing this physical inspection task, so AI cost is effectively infinite relative to human labor for this specific check. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | No production systems currently perform this task autonomously on aircraft. Research-stage computer vision systems exist but lack the robustness, integration, and certification required for safety-critical aviation use. Partial automation (e.g., barcode scanning assistance) exists but does not replace the full verification task. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product performs physical cabin supply and safety equipment checks; this remains a manual human task in commercial aviation today. |
Greet passengers boarding aircraft and direct them to assigned seats.
9CI 5–14 · exposure 8 · augmentation 13 · importance 4.0/5 · click for rater detail
Greet passengers boarding aircraft and direct them to assigned seats.
9| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Airlines operate in a highly regulated environment with strong unions and safety culture; there is no measurable production adoption of AI/robotic systems for passenger greeting and seating, and regulatory and labor barriers make rapid adoption unlikely. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | The airline cabin crew sector shows essentially no movement toward automating physical passenger greeting/seating tasks; this is a low-digitization, physical-service context. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | AI could assist by providing real-time seat availability or passenger information to attendants, but the core interpersonal and physical tasks of greeting and directing offer limited augmentation value for a human who is already present and performing the task. |
| Augmentation potential | claude-sonnet-5 | 1/5 | AI offers negligible assistance for the act of physically greeting and directing passengers to seats, though unrelated tools (seat maps, apps) exist outside this specific task. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI could verbally greet passengers, directing them to assigned seats requires real-time understanding of aircraft layout, passenger mobility needs, seating charts, and safe crowd management—tasks demanding spatial reasoning and human-like interaction that current systems cannot reliably execute end-to-end without human supervision. |
| Task automatability | claude-sonnet-5 | 1/5 | This requires physical presence, real-time human interaction, and identity/seat verification in a physical cabin space that current AI cannot perform end-to-end.The task is fundamentally physical and interpersonal, not amenable to software automation. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Safety regulations, customer expectations for human interaction, liability for misdirection or accidents, and FAA requirements for authorized crew to manage boarding create strong legal and organizational barriers to full automation. |
| Adoption barriers | claude-sonnet-5 | 4/5 | Aviation safety regulations require certified flight attendants physically present in the cabin for boarding and safety duties, creating strong regulatory and safety-driven barriers to automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | Building and deploying a system to greet passengers and navigate them to seats (whether robotic or multi-modal AI) would cost substantially more than a human flight attendant wage, and requires continuous oversight. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no viable AI substitute performing this physical task, so any comparison favors the human worker who is already embedded in required cabin staffing. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed product reliably performs physical greeting and seat-direction tasks in real aircraft environments; this requires embodied agents or sophisticated robotic systems not in production use today, well beyond current LLM or vision-only capabilities. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product performs physical passenger greeting and seating direction on aircraft; this remains firmly a human physical-presence task. |
Heat and serve prepared foods.
9CI 5–14 · exposure 8 · augmentation 13 · importance 3.4/5 · click for rater detail
Heat and serve prepared foods.
9| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Aviation is a highly regulated, conservative sector with strong labor agreements and safety constraints; adoption of autonomous food service automation is negligible, with no measurable production deployment in commercial operations. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Airlines have shown negligible movement toward automating physical cabin service tasks; this remains a manual, human-performed function with no robotic deployment trend. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | AI could modestly assist by predicting passenger meal preferences or optimizing galley workflow planning, but current systems offer minimal real-time support for the physical heating and serving activities themselves, limiting meaningful productivity augmentation. |
| Augmentation potential | claude-sonnet-5 | 1/5 | AI offers no meaningful assistance for the physical act of heating and serving food in an aircraft galley. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | Heating and serving prepared foods on aircraft involves physical handling of hot items, interaction with varied food storage systems, navigation of narrow aisles, and real-time adaptation to passenger needs. While AI could theoretically heat food (microwave/oven operation), the full task including transport on moving aircraft, safe serving to diverse passengers, and managing temperature-sensitive items remains largely manual and context-dependent, falling well short of 50% time savings at equal quality. |
| Task automatability | claude-sonnet-5 | 1/5 | This is a physical task requiring a human to be present in a moving aircraft cabin, operating galley equipment and serving passengers directly; no current AI system can perform physical heating and serving of food.','no software can accomplish this.' |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Multiple hard barriers exist: flight attendants are required by aviation regulation (FAA, EASA, etc.) to manage cabin service for safety and liability reasons; passenger preference for human interaction during service; strict aircraft certification and safety standards that would require extensive regulatory approval for any autonomous food-handling system; and the human-contact requirement inherent in hospitality and emergency-response duties. |
| Adoption barriers | claude-sonnet-5 | 4/5 | FAA/aviation safety regulations require certified flight attendants onboard for safety and service duties, and physical robotics in aircraft cabins face major safety, certification, and space constraints. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | The cost of robotics capable of safely heating, transporting, and serving food in an aircraft cabin—combined with integration, maintenance, and safety certification—would far exceed the loaded wage of a flight attendant performing this task over relevant time horizons. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no AI-based substitute for this physical service task, so AI cost is effectively infinite relative to the human performing it manually. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed commercial products currently perform end-to-end heating and serving of food on aircraft. Robotic food service is nascent and remains confined to laboratory settings or narrow, controlled environments; the dynamic, safety-critical nature of aircraft service precludes current-generation automation. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product exists to physically heat and serve food on aircraft; this remains purely a manual, physical-labor task performed by cabin crew. |
Reassure passengers when situations, such as turbulence, are encountered.
5CI 5–5 · exposure 0 · augmentation 13 · importance 4.2/5 · click for rater detail
Reassure passengers when situations, such as turbulence, are encountered.
5| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Aviation is a heavily regulated, safety-critical sector where passenger-facing automation adoption is extremely slow. There is no measurable trend toward AI replacement of flight attendants for reassurance tasks; regulatory and safety culture strongly favor human crew. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Airlines are not deploying AI to replace in-person passenger reassurance during flight; this remains a fully human-staffed safety function with no visible automation trend. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | AI could marginally assist by suggesting talking points or monitoring passenger distress signals, but the core task—delivering genuine human reassurance—is not substantially augmented by current AI tools, and flight attendants have limited opportunity to consult systems mid-crisis. |
| Augmentation potential | claude-sonnet-5 | 1/5 | AI offers essentially no in-the-moment assistance for a flight attendant calming passengers during turbulence, as this is an immediate interpersonal and physical-presence task. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | Reassuring passengers during turbulence requires genuine human empathy, social presence, and contextual emotional intelligence. Current AI systems cannot replicate the interpersonal trust and nonverbal communication that make reassurance effective in this high-stakes, emotionally charged scenario. |
| Task automatability | claude-sonnet-5 | 1/5 | Reassuring passengers during in-flight events like turbulence requires physical presence, human warmth, and real-time trust-building that current AI cannot replicate or perform end-to-end.' |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Federal aviation regulations (FAA/EASA) mandate flight attendants for safety; reassuring passengers is part of crew duty and passenger safety oversight. Human crew presence is legally required, and airlines face liability if they substitute AI for human judgment in passenger distress situations. |
| Adoption barriers | claude-sonnet-5 | 4/5 | Flight attendants are required by aviation safety regulations to be physically present and manage passenger safety and behavior, creating strong regulatory and physical-presence barriers to substitution. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | Even if an AI system could theoretically deliver reassurance, deployment cost (hardware, integration, maintenance) combined with inevitable human oversight would far exceed the marginal cost of having a flight attendant—who provides multiple other safety and service functions—perform this task. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no AI substitute performing this task, so cost comparison favors the human by default since AI cannot deliver the output at all. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed product reliably performs human-level passenger reassurance during turbulence. While chatbots exist, they lack the embodied presence, real-time situational awareness, and emotional credibility required to calm anxious passengers in a cabin environment. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product performs in-cabin passenger reassurance; this remains entirely a human interpersonal task with no automation in production. |
Conduct periodic trips through the cabin to ensure passenger comfort and to distribute reading material, headphones, pillows, playing cards, and blankets.
3CI 0–5 · exposure 0 · augmentation 13 · importance 3.8/5 · click for rater detail
Conduct periodic trips through the cabin to ensure passenger comfort and to distribute reading material, headphones, pillows, playing cards, and blankets.
3| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Airlines have shown minimal adoption of automation for in-cabin service; the sector remains highly conservative on crewed safety-critical roles and passenger interaction. No meaningful pilots or deployments exist in the airline industry for this type of service automation. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Airline cabin service is a highly physical, safety-regulated environment with essentially no AI/robotic adoption for these tasks. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | AI could assist with inventory tracking or recommending when items need restocking, but the core physical service task offers limited augmentation potential while the human performs it; the value is primarily in the direct human-passenger interaction and physical distribution itself. |
| Augmentation potential | claude-sonnet-5 | 1/5 | AI offers no meaningful assistance for the physical act of walking the cabin and distributing items to passengers. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | This task requires physical mobility through a confined aircraft cabin, direct social interaction with diverse passengers, and real-time judgment about comfort needs—capabilities far beyond current robotics or AI systems. No deployed technology can autonomously navigate cabin aisles, assess individual comfort states, and distribute physical items at scale. |
| Task automatability | claude-sonnet-5 | 1/5 | This requires physical presence, mobility through a cabin, manual handling of items, and personal interaction with passengers—none of which current AI systems can perform without a robotic embodiment far beyond deployed capability. |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Aviation is one of the most heavily regulated industries; aircraft operations require FAA certification, and any autonomous system performing passenger-facing duties would face stringent liability requirements, safety mandates, and the explicit need for human crew to remain present for emergency response. |
| Adoption barriers | claude-sonnet-5 | 4/5 | FAA regulations require certified flight attendants onboard for safety and service reasons, and airlines have strong operational and safety-related reasons to keep humans performing cabin duties. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | The hardware cost of a mobile robot capable of safe cabin operation, plus integration, maintenance, and liability insurance, would far exceed the loaded wage of a flight attendant for years. Physical automation in constrained spaces remains prohibitively expensive. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no viable AI substitute performing this physical task, so any AI-based alternative (e.g., robotics) would be vastly more expensive than a human flight attendant today. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No production systems exist that can physically perform in-cabin service tasks. While robots exist in lab settings, none are deployed on commercial aircraft for this purpose, and regulatory/safety barriers prevent such deployment today. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product performs physical in-cabin service and distribution tasks; this remains purely a human physical labor task. |
Assist passengers in placing carry-on luggage in overhead, garment, or under-seat storage.
3CI 0–5 · exposure 0 · augmentation 0 · importance 3.3/5 · click for rater detail
Assist passengers in placing carry-on luggage in overhead, garment, or under-seat storage.
3| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Aviation is a highly regulated, conservative sector with strong barriers to equipment changes. No measurable adoption of automated luggage-placement systems is evident in commercial fleets, and regulatory approval timelines are lengthy. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Airlines and aviation broadly show negligible adoption of physical robotic assistance for cabin tasks; this is a low-digitization, physical-labor task with no momentum toward automation. |
| Augmentation potential | claude-haiku-4-5-20251001 | 1/5 | Current AI offers no meaningful assistance to flight attendants performing this physical task; there is no augmentation pathway for a human working in the loop with today's technology. |
| Augmentation potential | claude-sonnet-5 | 1/5 | AI offers no meaningful assistance to a flight attendant physically helping a passenger lift and stow luggage in an overhead bin. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | This task fundamentally requires physical manipulation of objects in a constrained space and real-time interaction with passengers, which current AI systems cannot perform. No deployed robotic system reliably handles variable luggage sizes, weights, and passenger needs in the aircraft cabin environment. |
| Task automatability | claude-sonnet-5 | 1/5 | This requires physical manipulation of luggage and direct physical assistance to passengers, which no current AI system (embodied or otherwise) can perform in an aircraft cabin setting. |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Significant legal and safety barriers exist: aviation regulators require human supervision in safety-critical cabin environments, liability for passenger injury or luggage damage falls on the airline, and there is an explicit human-contact requirement for passenger assistance and safety assurance. |
| Adoption barriers | claude-sonnet-5 | 4/5 | Beyond physical feasibility, safety regulations require certified crew to manage cabin safety including luggage stowage, and liability for injury during boarding assistance is a serious concern favoring trained humans. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | The cost of deploying a robotic system capable of safely handling luggage placement (hardware, maintenance, integration, safety certification) would far exceed the hourly wage of flight attendants for this specific task component. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no viable AI/robotic substitute at any cost point for this physical task in the cramped, dynamic cabin environment, making the human by far the cheaper and only functional option. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No production system exists that can autonomously place carry-on luggage in overhead or under-seat storage. This requires embodied robotics capable of safe, precise object handling in confined spaces with human safety constraints—not yet deployed in commercial aviation. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed robotic or AI product assists passengers with carry-on luggage stowage in commercial aviation; this remains entirely a research-stage robotics challenge if attempted at all. |
Sell alcoholic beverages to passengers.
3CI 0–5 · exposure 0 · augmentation 25 · importance 2.9/5 · click for rater detail
Sell alcoholic beverages to passengers.
3| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Aviation is a heavily regulated, safety-critical sector with slow AI adoption in passenger-facing operations. No measurable deployment of automated beverage service exists in commercial airlines. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Airlines are a highly physical, safety-regulated service sector with essentially no movement toward automating in-cabin beverage service. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | Digital ordering systems or inventory management tools could modestly assist attendants, but core sales activities—face-to-face interaction, persuasion, payment handling—remain primarily human-dependent. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI could assist with inventory tracking or point-of-sale payment processing, but offers minimal enhancement to the core physical selling interaction. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | Selling beverages requires real-time interaction with passengers, taking payment, handling cash/cards, and navigating social dynamics in a confined space. Current AI cannot operate independently in this dynamic, in-person environment. |
| Task automatability | claude-sonnet-5 | 1/5 | This requires physical presence, handling cash/cards, checking ID, and serving items in the cabin—no current AI system can physically perform this task. |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Federal aviation regulations and airline policies require human crew members for in-flight service and safety compliance. Human presence is legally mandated on flights, creating hard barriers to automation. |
| Adoption barriers | claude-sonnet-5 | 4/5 | Safety regulations, alcohol service liability/age verification laws, and FAA crew requirements mean a certified human flight attendant must handle this, creating strong regulatory and liability barriers. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | AI systems capable of in-cabin beverage service would require robotics, payment integration, and oversight costs that vastly exceed the loaded wage of a flight attendant performing this task. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no AI substitute performing this physical service, so AI cost comparison is not applicable and effectively the human is required, making AI more 'expensive' in the sense of infeasible. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed AI system can reliably perform beverage sales on aircraft today. The task requires physical presence, payment processing, and passenger judgment that deployed products do not demonstrate. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed AI product performs in-flight physical beverage sales; this is purely a human physical service task. |
Announce and demonstrate safety and emergency procedures, such as the use of oxygen masks, seat belts, and life jackets.
1CI 0–3 · exposure 0 · augmentation 25 · importance 4.8/5 · click for rater detail
Announce and demonstrate safety and emergency procedures, such as the use of oxygen masks, seat belts, and life jackets.
1| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Aviation is heavily regulated; safety procedures are non-negotiable and mandated by law. There is no adoption of AI replacement in this domain, and regulatory barriers prevent meaningful substitution. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Airlines have adopted recorded safety videos for decades but the live crew requirement remains unchanged; the sector shows slow structural change in this specific task despite tech-forward practices elsewhere in aviation. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | AI could assist with multimedia content (animations, translations, or real-time captioning) during demonstrations, but the human flight attendant remains the required performer. Assistance value is limited given the straightforward, standardized nature of the task. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI could help produce or personalize safety videos or translate announcements, but it offers minimal augmentation to the attendant's live physical demonstration and passenger safety monitoring duties. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | Safety demonstrations require real-time human presence, attention management, and responsive interaction with passengers in a physical cabin environment. Current AI cannot perform this task end-to-end; it cannot physically don and demonstrate equipment, respond to passenger questions, or command the social authority needed for safety compliance. |
| Task automatability | claude-sonnet-5 | 1/5 | This requires physical presence in the cabin to demonstrate equipment and manage passenger attention, which cannot be performed end-to-end by current AI systems.}, |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Federal aviation regulations (FAA Part 121) explicitly require a qualified flight attendant to conduct safety demonstrations on each flight. This is a hard legal barrier—a licensed human must perform or supervise the task. |
| Adoption barriers | claude-sonnet-5 | 5/5 | Aviation safety regulations (FAA/EASA etc.) mandate a certified crew member perform and be present for safety briefings and emergency procedures, making this a hard legal/regulatory barrier. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | Deploying robotic systems or enhanced video systems to replace a flight attendant's safety demonstration would cost far more than the attendant's loaded wage, especially given integration, maintenance, and regulatory compliance costs. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no AI system performing this physical demonstration task, so no viable cost comparison to a human favors AI; any automation (video systems) already exists as a low-cost supplement, not an AI replacement of the person. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed AI system performs in-cabin safety demonstrations in production. Video playback systems exist but do not replace the live human-led demonstration, which is mandated by regulation and serves as a critical attention and compliance mechanism. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product substitutes for a physically present crew member demonstrating safety equipment; pre-recorded videos exist but the flight attendant's live announcement/monitoring role remains a separate, undelegated task. |
Verify that first aid kits and other emergency equipment, including fire extinguishers and oxygen bottles, are in working order.
0CI 0–0 · exposure 0 · augmentation 25 · importance 4.9/5 · click for rater detail
Verify that first aid kits and other emergency equipment, including fire extinguishers and oxygen bottles, are in working order.
0| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Aviation safety protocols are heavily regulated and change slowly; airlines have no incentive or regulatory pathway to automate crew-performed safety inspections, limiting any adoption even conceptually. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Airlines are a highly regulated, safety-critical physical-world sector where this specific physical safety verification task shows no adoption of AI automation. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | AI might assist via checklist prompts or image recognition of obviously degraded equipment (e.g., dents, visible corrosion), but the core task of hands-on verification and sign-off must remain with the human attendant, so augmentation is minimal. |
| Augmentation potential | claude-sonnet-5 | 2/5 | Digital checklists or IoT sensors could theoretically flag equipment status, but current AI offers minimal meaningful assistance to the hands-on physical verification process itself. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | This task requires hands-on physical inspection and tactile verification of equipment condition, functional testing (e.g., checking pressure gauges, verifying seals), and professional judgment about safety compliance—all impossible for current AI systems to perform end-to-end without human presence. |
| Task automatability | claude-sonnet-5 | 1/5 | This requires physical presence to visually inspect, check gauges, and verify expiration dates on physical equipment in an aircraft cabin—no current AI system can perform this physical verification task. |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Federal aviation regulations (FAA/EASA) mandate that flight crews perform and document these safety checks; liability and passenger safety are tightly regulated, making human authorization legally required and non-delegable to automation. |
| Adoption barriers | claude-sonnet-5 | 5/5 | Aviation safety regulations (FAA/EASA) mandate certified flight attendants perform and document these safety checks, creating hard legal and liability barriers to automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | AI has no practical cost advantage here because the task cannot be meaningfully automated—human flight attendants must physically inspect these items as part of pre-flight duties, so comparison is moot. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no AI substitute for physical inspection, so cost comparison favors the human who must physically be present and perform the check; AI cannot replace this at any cost currently. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed AI product can independently verify emergency equipment integrity; this task demands in-person visual inspection, manual checks, and safety sign-off that no current system performs in production aviation environments. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed AI product performs physical safety equipment checks on aircraft; this remains a manual human inspection task mandated by aviation safety protocols. |
Monitor passenger behavior to identify threats to the safety of the crew and other passengers.
0CI 0–0 · exposure 0 · augmentation 25 · importance 4.7/5 · click for rater detail
Monitor passenger behavior to identify threats to the safety of the crew and other passengers.
0| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Aviation is a highly regulated, risk-averse sector with slow technology adoption cycles for safety-critical tasks. No measurable adoption of AI-based passenger threat detection exists in commercial operations, and regulatory constraints strongly inhibit deployment velocity. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Airlines are a highly regulated, safety critical physical-world sector with minimal AI adoption for in-cabin passenger monitoring; this remains firmly human-performed. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | While AI could theoretically flag visual anomalies or alert systems, the core task requires human judgment, cultural sensitivity, and real-time communication skills that AI augmentation cannot meaningfully enhance; human attendance remains essential and cannot be displaced even partially. |
| Augmentation potential | claude-sonnet-5 | 2/5 | Some cabin sensor or camera systems could flag anomalies for attendants, but this is not a mature or widespread capability and adds only marginal assistance today. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | Monitoring passenger behavior for safety threats requires real-time situational awareness, nuanced interpretation of context and body language, and immediate intervention or communication with crew—capabilities far beyond current AI systems operating on aircraft. No end-to-end automation exists that could reduce task time by 50% while maintaining safety standards. |
| Task automatability | claude-sonnet-5 | 1/5 | Real-time interpretation of ambiguous human behavior in a dynamic cabin environment requires physical presence, situational judgment, and immediate physical intervention capability that current AI cannot replicate., especially without embodiment. |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Federal Aviation Administration (FAA) regulations mandate that flight attendants perform safety duties, and crew monitoring of passengers is a legally required function tied to air-crew certification and liability. Automation of this task faces hard regulatory barriers requiring licensed human responsibility. |
| Adoption barriers | claude-sonnet-5 | 5/5 | Aviation safety regulations mandate certified crew members to perform in-cabin safety and security monitoring, and liability for missed threats is severe, making regulatory and legal barriers to substitution very high. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | Integration of monitoring AI on aircraft would require specialized hardware, continuous model refinement, regulatory certification, and human oversight systems that would exceed the cost of existing flight attendant labor, particularly given liability and safety margins required. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no viable AI substitute performing this function, so no meaningful cost comparison exists; any hypothetical system would require extensive sensors, certification, and oversight making it costlier than a trained employee. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | While computer vision systems exist for general surveillance, deploying them to detect subtle behavioral threats in dynamic cabin environments with sufficient reliability for safety-critical decisions is not demonstrated in production on commercial flights. Current technology lacks the contextual reasoning and liability-safe error rates needed. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product performs onboard passenger behavioral threat monitoring in place of a human flight attendant; camera-based anomaly detection remains experimental and unused for this function. |
Walk aisles of planes to verify that passengers have complied with federal regulations prior to takeoffs and landings.
0CI 0–0 · exposure 0 · augmentation 13 · importance 4.7/5 · click for rater detail
Walk aisles of planes to verify that passengers have complied with federal regulations prior to takeoffs and landings.
0| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Regulatory mandates and safety-critical nature mean aviation is one of the most conservative sectors for automation. No measurable pilot or production adoption of automation for pre-flight cabin safety checks exists. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Aviation safety-critical physical tasks show essentially no AI/robotic adoption; the sector relies on human crew for these functions with no displacement trend. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | Augmentation is minimal because the task is already highly visual and procedural; flight attendants cannot be absent from the cabin during compliance checks. Digital checklists or alerts might slightly streamline reporting, but do not meaningfully transform productivity. |
| Augmentation potential | claude-sonnet-5 | 1/5 | AI offers no meaningful assistance to the physical act of walking aisles and visually verifying compliance; this is a hands-on human function with no current tool support. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | This task requires physical movement through the plane cabin, visual inspection of diverse passenger scenarios, and real-time judgment about regulatory compliance. Current AI systems lack the embodied capability to walk aisles and cannot reliably assess human behavior variations in an unstructured environment. |
| Task automatability | claude-sonnet-5 | 1/5 | This requires physical presence, mobility through a cabin, and real-time visual/physical verification of passenger compliance (seatbelts, tray tables, luggage), which current AI cannot perform end-to-end without embodiment.' |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Federal Aviation Regulations (FAA Part 121) legally mandate that flight attendants—human crew members—conduct these safety checks before takeoff and landing. This is a hard legal and safety requirement that cannot be delegated to automated systems. |
| Adoption barriers | claude-sonnet-5 | 5/5 | FAA regulations explicitly require certificated flight attendants to perform safety checks; this is a hard legal and safety-critical requirement enforced by regulation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | Any hypothetical automation system would require extensive hardware (mobile robots, distributed sensors, integration with cabin architecture) whose development and deployment cost far exceeds the loaded wage of flight attendants performing this task. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no viable AI alternative performing this physical task, so any hypothetical automation (e.g., robotics) would be far more costly than a human flight attendant today. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed product performs this task in production. While computer vision could theoretically identify seatbelts or upright seats in limited scenarios, end-to-end cabin compliance verification with legal liability remains strictly human-operated in all commercial aviation operations. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product performs physical cabin walk-throughs and compliance verification; this remains a physical human task with no robotic or AI substitute in production. |
Direct and assist passengers in emergency procedures, such as evacuating a plane following an emergency landing.
0CI 0–0 · exposure 0 · augmentation 25 · importance 4.7/5 · click for rater detail
Direct and assist passengers in emergency procedures, such as evacuating a plane following an emergency landing.
0| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Regulatory requirements and safety-critical nature mean this task has zero adoption of automation and will not change in any foreseeable timeframe. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Airlines and aviation safety regulators show essentially no movement toward automating physical emergency evacuation assistance; this is a highly regulated, safety-critical physical function. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | While AI could potentially provide real-time decision support or communication aids during emergencies, the core task of physically directing passengers must remain human-led, limiting meaningful augmentation potential. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI can support training simulations, emergency checklists, or communication systems, but offers minimal real-time assistance during an actual physical evacuation event. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | Emergency evacuation direction requires real-time human judgment, physical presence, and the ability to make split-second decisions under extreme stress based on dynamic conditions that AI cannot perceive or influence in an aircraft cabin. |
| Task automatability | claude-sonnet-5 | 1/5 | This is a physical, safety-critical task requiring real-time human judgment, physical presence, and hands-on assistance during chaotic emergencies; no AI system can perform this end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Federal aviation regulations (FAA, EASA, ICAO) explicitly mandate trained flight attendants on commercial aircraft to direct emergency procedures; no automation or delegation is legally permitted. |
| Adoption barriers | claude-sonnet-5 | 5/5 | Aviation safety regulations (e.g., FAA/EASA) mandate certified human flight attendants to perform emergency evacuation duties, and liability, training, and certification requirements are extremely strict. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | Flight attendants are legally required personnel whose primary safety function cannot be replaced by automation; the cost comparison is irrelevant since human presence is mandatory. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no AI substitute performing this physical task, so cost comparison is moot; humans remain the only viable option, making AI effectively infinitely costlier or inapplicable. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed AI system can physically direct passengers, assess cabin conditions, or command authority during emergencies; this task inherently requires a human agent present on the aircraft. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product performs physical emergency evacuation direction and passenger assistance; this remains entirely a human function performed by trained crew. |
Prepare passengers and aircraft for landing, following procedures.
0CI 0–0 · exposure 0 · augmentation 25 · importance 4.7/5 · click for rater detail
Prepare passengers and aircraft for landing, following procedures.
0| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Aviation is heavily regulated and safety-critical; there is no adoption trend toward automation of in-cabin landing procedures because regulatory and safety barriers make it infeasible. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | The airline cabin safety sector shows essentially no movement toward automating physical passenger safety procedures; adoption in this specific task domain is negligible. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | AI could potentially assist with checklist management or real-time communication systems, but the core task—physically preparing passengers and securing the cabin—requires direct human performance with minimal meaningful augmentation opportunities. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI could marginally assist with checklists or training simulations, but it offers little direct in-the-moment assistance during actual physical landing preparation duties. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | Landing preparation involves physical presence with passengers (safety instructions, securing cabin items, securing oneself), customer interaction, and real-time assessment of cabin conditions—tasks that fundamentally require human agency and physical presence in the aircraft cabin. |
| Task automatability | claude-sonnet-5 | 1/5 | This task requires physical presence in a cabin, direct human interaction with passengers, physical safety checks (seatbelts, tray tables, doors), and hands-on emergency preparedness that current AI cannot perform end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Federal aviation regulations (FAA, EASA) explicitly mandate that flight attendants must be present and perform safety duties during landing; this is a hard legal and safety requirement that cannot be bypassed. |
| Adoption barriers | claude-sonnet-5 | 5/5 | Aviation regulations (FAA and international equivalents) mandate certified human flight attendants to perform safety duties including landing preparation, making automation legally prohibited. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | The task requires human physical presence and safety responsibility; no AI system can substitute for the loaded cost of a flight attendant, as no viable automation exists. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no AI substitute performing this physical, safety-critical task, so cost comparison favors the human by default since AI cannot deliver the output at all. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed AI system can physically perform the in-cabin actions, passenger interactions, and safety verifications required during landing preparation; this remains outside the scope of current autonomous systems. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed AI product performs passenger safety briefings, physical cabin checks, or hands-on landing preparation; this remains entirely a human physical and social task. |
Administer first aid to passengers in distress.
0CI 0–0 · exposure 0 · augmentation 25 · importance 4.6/5 · click for rater detail
Administer first aid to passengers in distress.
0| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | There is no adoption of AI for this task because the task cannot be performed by machines and is mandated by regulation to remain human-delivered. No pilots, production deployments, or displacement occur. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Airline cabin crew safety functions are minimally digitized for physical intervention tasks, and there is no trend toward automating hands-on emergency care in this sector. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | AI could provide limited assistance (e.g., AI-guided step-by-step prompts on a tablet, symptom checkers, or telemedicine consultation routing) but the core manual and decision-making load remains on the flight attendant; most first-aid tasks in flight require tactile and real-time judgment with minimal scope for significant productivity gain from current assistive tools. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI-enabled decision-support tools (e.g., symptom-checker apps or remote medical consultation via satellite link) can provide guidance, but the core physical task is untouched by AI assistance. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | First aid to passengers in distress requires real-time physical intervention, dynamic clinical decision-making under uncertainty, and immediate physical presence on an aircraft—none of which current AI can perform. The task is inherently embodied and contextual in ways that preclude end-to-end automation today. |
| Task automatability | claude-sonnet-5 | 1/5 | Physical administration of first aid to a distressed passenger requires hands-on human presence, dexterity, and real-time judgment that current AI cannot perform end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Hard legal and safety barriers exist: aviation regulations (FAA, EASA, etc.) and airline protocols require a trained human crew member to deliver first aid and make clinical judgments; liability and safety-of-life requirements make autonomous substitution infeasible and unlawful. |
| Adoption barriers | claude-sonnet-5 | 5/5 | Emergency medical response onboard aircraft is governed by aviation safety regulations and crew certification requirements, and liability for medical care mandates trained human responders. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | AI cannot perform this task at all, so the cost ratio is undefined; a human flight attendant with first-aid training is mandatory. There is no substitution pathway. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no AI system capable of substituting for this task, so cost comparison favors the human by default; any AI role is a marginal decision-support add-on. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed AI system can physically deliver first aid, assess vital signs through touch/observation in flight conditions, or adapt treatment in real time to a specific passenger's physiology and medical history. This remains a purely human-performed task in production. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product performs physical first aid autonomously; this remains purely a human physical-response task with no robotic substitute in aviation cabins. |
Assist passengers entering or disembarking the aircraft.
0CI 0–0 · exposure 0 · augmentation 0 · importance 3.9/5 · click for rater detail
Assist passengers entering or disembarking the aircraft.
0| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | This task occurs in a highly regulated, safety-critical sector (aviation) where human presence and accountability are mandated; no measurable adoption of AI or robotics for in-cabin passenger assistance is occurring. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Airline cabin crew operations show negligible AI adoption for physical passenger handling tasks, as this remains a manual, safety-regulated function. |
| Augmentation potential | claude-haiku-4-5-20251001 | 1/5 | AI offers minimal assistive value; the task is inherently hands-on and requires immediate physical response that a human attendant must provide. Informational systems (e.g., alerts) are peripheral to the core assistance function. |
| Augmentation potential | claude-sonnet-5 | 1/5 | AI offers little to no direct assistance for the physical act of helping passengers board or disembark, though unrelated tools like scheduling software don't apply to this specific task. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | This task fundamentally requires physical presence, real-time navigation of a dynamic, narrow space, and responsive interaction with humans at varying levels of mobility and need. Current AI systems cannot physically guide, support, or stabilize passengers. |
| Task automatability | claude-sonnet-5 | 1/5 | This is a physical, in-person task requiring real-time human assistance (helping with luggage, mobility issues, seating) that no current AI or robotic system can perform end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Strong legal, safety, and liability barriers exist: passengers' physical safety during boarding/deplaning requires a qualified human present; airlines face duty-of-care obligations and liability if an automated system fails to assist a mobility-impaired or elderly passenger appropriately. Regulatory oversight of aviation safety is extensive. |
| Adoption barriers | claude-sonnet-5 | 5/5 | Safety regulations (FAA/EASA) require certified flight attendants to be physically present and perform passenger safety and boarding duties, making this a hard regulatory and physical barrier. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | A robot capable of physically assisting passengers would cost far more to develop, maintain, and deploy than the loaded wage of a flight attendant performing the task today. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no AI substitute performing this physical task, so cost comparison favors the human by default since no viable AI alternative exists. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed AI product performs this task. Humanoid robots capable of safely assisting passengers in aircraft cabins do not exist in production; the task demands dexterity, balance, and real-time judgment in a constrained environment. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product performs physical passenger boarding assistance; this remains entirely a human physical-presence task in commercial aviation. |
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.