EXPOSED
This is a residual catch-all — companions and sitters, wedding and funeral attendants, chaperones, bath and locker room attendants, doula-adjacent and non-medical support roles — and the modal worker spends the day physically present with a client, escorting, assisting, supervising, and reassuring. Almost none of that is text or screen work AI can take, and no robot can walk a frail client to the bathroom or calm a nervous bride. What is exposed is the coordination layer around the work: scheduling, client matching, intake paperwork, activity planning, and the platform middlemen who currently connect these workers to clients.
Most of this decline happened after 2021 — it is not the pandemic dip.
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.
BLS projection, 2024–2034
+6.4%
Percentage only. The projection counts a different population from the 60,420 above — it includes self-employed workers, which for this occupation is most of them, so the two headcounts are not comparable.
Growing, and only partly exposed
The BLS expects +6.4% more of these jobs by 2034, and at 54/100 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.
~16,100 openings a year on average, including replacing people who leave.
DoulaValetButlerEscortDresserFootmanGreeterServantChaperonCrematorTattooerBootblackChaperoneShoeblackTattooistDog WalkerDoorkeeperShoeshinerShoe ShinerTaxi DancerBody PiercerBrush WorkerHouse SitterShine Worker
The BLS uses Personal Care and Service Workers, All Other for work that doesn't fit any named occupation, so it covers roles that have little in common with each other. Two consequences worth knowing before you read anything below:
Holding it up: embodiment . Weakest point: liability shield .
Tasks largely resist digitisation Walking a frail companion client to the toilet, holding a bride's train at the church door, or staying awake overnight as a sitter are tasks no software executes — 15 rather than 18 because the paperwork these workers do carry (intake forms, activity logs, shift notes, mileage and hour reporting) plus the client-matching and scheduling that fills the gaps is already being absorbed by apps.
Hands-on in uncontrolled environments The work happens in other people's homes, hotel ballrooms, funeral chapels, locker rooms and pool decks — uncontrolled spaces with wet floors, unpredictable stairs, agitated or grieving people, and physical contact from a steadying arm to lifting a wheelchair footrest, which is why this sits at 16 and not higher only because it rarely involves heavy machinery or clinical procedures.
No licence, no signature requirement Companion and sitter work is specifically carved out of most state nurse-practice acts precisely because no licence is required — a background check, CPR card, and sometimes a food-handler or pool-attendant permit is the ceiling, so a 3 reflects screening, not credentialing, and the agency's insurance carries the exposure rather than the worker's name.
Meaningful discretion Real calls happen — deciding a client is confused enough to phone the family, deciding a chaperoned teenager needs to be separated from a group, deciding a guest is too intoxicated to drive from the reception — but the standing instruction is escalate to the agency, the nurse, or the family, so at 8 the discretion is about recognising a situation, not owning the resolution.
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 (15/20 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 (3/20) is whether the law requires a licensed human to sign. Trust premium (12/20) is whether buyers specifically pay for a person. Judgment and accountability (8/20) 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 23 of this occupation's 54 points (43%).
Embodiment (16/20) 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.
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.
Doula licensure/Medicaid reimbursement pathways already enacted in ~a dozen states (e.g. Oregon, Minnesota, Rhode Island, Michigan) create a credentialed, enrollable provider with a scope of practice. This only lifts the sub-segment of the residual code that is doula-adjacent, not chaperones or locker room attendants.
State-level certification/registry mandates for non-medical companion and in-home care workers tied to Medicaid HCBS billing — e.g. the CMS Access Rule's 80/20 direct-care compensation provision plus state background-check and training registries (Washington's HCA certification for long-term care workers is the closest existing model). If a state requires a registered individual to be named on the service log for reimbursement, and that individual carries mandatory-reporter duty for suspected abuse/neglect, the shield stops being nominal.
Task-mix shift plus formal escalation duty: once scheduling, intake and activity planning are automated, the residual day is fall risk, cognitive decline observation, and deciding when to call 911 or a family member. If state HCBS rules or agency contracts require the companion to document and escalate change-of-condition observations (already standard in some assisted-living aide protocols), the role owns a consequential call under ambiguity rather than merely reporting hours.
Buyers in this category are already paying for physical human presence, so the premium is not about anti-AI branding but about defection from platform intermediation: if Care.com/Papa-style matching layers are seen as extracting margin while AI handles matching, direct-hire and agency-of-record relationships where the family chooses a named individual raise the identified-human component. Watch for state domestic-worker bills of rights (NY, Seattle, Philadelphia) extending written-agreement requirements to companion care.
The limit. This is a residual SOC bucket, so levers apply unevenly: doula and companion-care segments have real licensure and Medicaid pathways in motion, while wedding/funeral attendants, chaperones and locker room attendants have essentially no route to a liability shield and their trust premium is wage-suppressed rather than credential-protected. Aggregate score gains would likely be diluted by the non-care half of the code, and the automatable coordination layer keeps downward pressure on hours even where the in-person task is safe.
| Portland-Vancouver-Hillsboro, OR-WA | 11,120 | $41,600 +0% |
| Salem, OR | 2,870 | $41,600 +0% |
| Eugene-Springfield, OR | 2,350 | $41,600 +0% |
| New York-Newark-Jersey City, NY-NJ | 1,860 | $42,000 +1% |
| Miami-Fort Lauderdale-West Palm Beach, FL | 1,690 | $34,830 -16% |
| Los Angeles-Long Beach-Anaheim, CA | 1,600 | $43,160 +4% |
| Atlanta-Sandy Springs-Roswell, GA | 1,440 | $23,850 -43% |
| Las Vegas-Henderson-North Las Vegas, NV | 1,280 | $34,960 -16% |
| Salinas, CA | 50 | $50,910 +22% |
| Norwich-New London-Willimantic, CT | 30 | $48,320 +16% |
| Charleston-North Charleston, SC | 70 | $47,320 +14% |
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 54. 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.