EXPOSED
Passenger attendants — on ferries, trams, transit shuttles, tour buses and paratransit — spend most of their shift physically present with riders: securing wheelchairs, helping elderly passengers board, checking fares, announcing stops, handling spills and disputes. Language AI barely touches that work, and robotics is nowhere near lifting a rider into a lift-equipped van. The real threat is not AI reasoning but automated fare gates, kiosks, onboard cameras and automated announcement systems that shrink headcount per vehicle, plus driverless transit designs that trim attendant roles to one per train rather than one per car.
Roughly flat across the period, with year-to-year wobble.
Median pay $26,570 → $37,720 +13.6% in real terms
This line is counted by the Bureau of Labor Statistics — the one figure on this page that isn't a judgement of ours. Headcount moves on demand, offshoring, demographics and the business cycle, and automation is one term among several, often not the loudest.
So a falling line is not evidence that AI did it, and a rising one is not evidence that it won't. Both happen in this register: some occupations resist automation and shrink anyway, others are highly automatable and keep growing. The marked year is 2020.
BLS projection, 2024–2034
+4.7% 25,600 → 26,800 on the projections basis
Growing, and only partly exposed
The BLS expects +4.7% more of these jobs by 2034, and at 47/100 the work is only partly exposed — some tasks are automatable, the core of the job is not. Nothing here is in tension.
Different clocks. The score is what current AI could do to this work today. The projection is how many of these jobs will exist in 2034. Everything between the two — how fast employers actually adopt, whether demand grows in the meantime — is why they can point opposite ways without either being wrong.
~4,100 openings a year on average, including replacing people who leave.
HostPurserHostessMonitorStewardBus AideAttendantBath AideCar PorterStewardessBus MonitorBus StewardFood PorterGround HostBath StewardReturn AgentShip StewardBus AssistantBus AttendantStation UsherDining Car HopGround HostessBath StewardessTrain Attendant
Holding it up: embodiment . Weakest point: liability shield .
Mixed — a routine tier and a judgment tier Securing a wheelchair with four-point tiedowns on a moving van, guiding a blind rider to a seat, and physically calming a dispute are not digitisable, but the tasks that fill the rest of the shift — fare validation, stop announcements, ticket scanning, headcount, incident logging — have already been replaced by tap gates, GPS-triggered PA and onboard cameras on most systems, which pulls it down from the high teens to 13.
Hands-on in uncontrolled environments You are on your feet in a vehicle in motion, kneeling on wet floors, working lift ramps in rain and ice, and bracing passengers on curbside grades — uncontrolled enough for a 15, short of 18+ only because the environment is a bounded vehicle interior rather than open terrain or an unpredictable worksite.
No licence, no signature requirement No state licence attaches to the role; a CPR/first-aid card, wheelchair-securement training and a transit agency background check are the whole gate, and when something goes wrong the FTA-regulated agency and the CDL-holding driver absorb the claim, not you — the 4 rather than 0 reflects that securement certification is contractually mandatory on ADA paratransit runs.
Executes defined procedures on defined inputs Most calls are scripted by agency policy: fare disputes go by posted rules, medical events trigger a radio call to dispatch and 911, evacuation follows a drilled sequence — the discretion you do exercise is real (deciding a rider is too impaired to board, when to stop the vehicle) but it is narrow and immediately escalated, which is why it sits at 6 rather than 10.
The verdict above describes this occupation as a whole. Almost nobody does the typical version of a job — tick what's actually in your week and see how your own mix sits.
Your task mix speaks to task resistance (13/20 here) — how much of the day's work current AI already does. That is the dimension the boxes above are about.
It cannot move the other three. Liability shield (4/20) is whether the law requires a licensed human to sign. Trust premium (9/20) is whether buyers specifically pay for a person. Judgment and accountability (6/20) is whether the role exists to own consequential calls. Those are facts about the occupation's standing, not about which tasks are in your week — a paralegal who does only trial exhibits still holds no licence. Together they are 19 of this occupation's 47 points (40%).
Embodiment (15/20) is also a property of the work rather than the worker, but we don't tag individual tasks as physical or not, so the picker can't tell you anything about it. That's a limit of this tool, not a claim.
Did we get the list right? Tell us what's missing — the tasks are written from the outside, and you're reading this from the inside.
Bus Drivers, School SAFE
The moves above are yours to make. This is the other half: what would have to change in the world for the occupation itself to score higher. None of it is in any one person's gift, but it is where the floor actually comes from. Scores here are not a one-way ratchet. Only two of the five dimensions — task resistance and embodiment — track what machines can do. The other three track law, what buyers will pay for, and who is answerable, and those move in both directions, often in response to the same pressure AI creates. If every lever below landed, this occupation would score around 68/100 — SAFE.
State DOT / FTA rules requiring a trained, certified attendant physically onboard whenever a driverless or highly automated transit vehicle carries passengers — the pattern in California DMV AV deployment permits and in NFPA/FTA State Safety Oversight plans for automated people movers, and explicitly in existing rail one-person-train-operation waivers. Also ADA-driven mandates: a rule that securement of a wheelchair in a paratransit or transit vehicle must be performed and attested by a certified attendant (PASS-certified, CTAA) rather than by the rider or an unsupervised system, with the agency's liability turning on that attestation.
Task-mix shift: fare collection, stop announcements and passenger counting are the automatable tier and are already largely gone on tap-to-pay and automated-announcement systems. What remains — lifting and securing riders, spills, aggressive passengers, cognitively impaired riders who miss stops — is the tier neither language AI nor robotics touches, so the residual job scores higher on this dimension even as headcount falls.
If headcount shrinks to one attendant per automated train or ferry, the remaining role absorbs evacuation command, medical-emergency triage, and de-escalation authority for the whole vehicle — an emergency-response role with documented decisions, rather than a fare-check role. Watch for FTA emergency-preparedness rules naming the onboard attendant as incident commander until responders arrive.
Union contracts and state law codifying minimum staffing per driverless train/ferry rather than per system — e.g. legislation of the type ATU and TWU have pushed (New York's two-person crew law for freight, similar transit bills in CA, IL) extended to automated transit, making an attendant a legally required presence for emergency evacuation certification.
Narrow route only: premium tour operators, cruise tenders and heritage rail already market a human attendant as part of the experience, and disability-rights advocacy (e.g. NCIL, DREDF comments on AV rulemaking) pushes agencies to guarantee a human helper on paratransit rather than an app. Broad transit riders do not pay extra for attendants, so this cannot rise far.
The limit. The threat here is headcount, not capability, and no lever reverses that. Even with strong staffing mandates the occupation shrinks: one attendant per driverless train instead of one per car is a 60-80% cut while every surviving role scores higher. A rising register score for this occupation is compatible with most of the 27,110 jobs disappearing. Trust premium is structurally capped — transit is a public budget line, and riders do not choose operators.
| New York-Newark-Jersey City, NY-NJ | 3,410 | $41,590 +10% |
| Houston-Pasadena-The Woodlands, TX | 1,780 | $31,240 -17% |
| Los Angeles-Long Beach-Anaheim, CA | 1,200 | $39,460 +5% |
| Boston-Cambridge-Newton, MA-NH | 1,150 | $43,350 +15% |
| Miami-Fort Lauderdale-West Palm Beach, FL | 990 | $31,750 -16% |
| Seattle-Tacoma-Bellevue, WA | 820 | $41,760 +11% |
| Washington-Arlington-Alexandria, DC-VA-MD-WV | 790 | $49,190 +30% |
| Denver-Aurora-Centennial, CO | 780 | $37,550 +0% |
| Washington-Arlington-Alexandria, DC-VA-MD-WV | 790 | $49,190 +30% |
| Cincinnati, OH-KY-IN | 70 | $44,810 +19% |
| San Francisco-Oakland-Fremont, CA | 510 | $44,800 +19% |
We have no reported case of a named organisation automating this occupation. Not one deployment, not one announcement.
That is worth saying out loud next to a score of 47. The verdict above is about what the work exposes — what current AI could do to these tasks. It is not a claim that anyone has done it. For this occupation those two things have come apart completely: the capability argument is on this page, and the evidence column is empty.
Has AI actually changed your work? One tap, anonymous, and the running tally is public. Nothing else is asked of you.
Rather than check back: get the digest and we'll tell you what changed — or watch a single occupation from its own page.