EXPOSED
This catch-all covers specialty cleaning work — chimney sweeping, duct and vent cleaning, pressure washing, high-rise window and exterior work, post-construction and biohazard cleanup — performed by hand in cramped, dirty, variable spaces that no current robot navigates. Language AI touches almost none of it; the exposure here is scheduling, quoting, and dispatch getting absorbed by software, plus commercial floor-scrubbing robots nibbling at the flat-open-space end. The score is middling not because the labor is automatable but because there is no license, no signature, and little client relationship protecting the wage.
This fall is concentrated in 2020 and has not recovered since.
Median pay $31,230 → $44,040 +12.8% 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
+2.5% 18,100 → 18,500 on the projections basis
Growing, and only partly exposed
The BLS expects +2.5% more of these jobs by 2034, and at 55/100 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.
~2,600 openings a year on average, including replacing people who leave.
Rug CleanerDuct CleanerChimney SweepChimney SweeperHand Rug CleanerTree Trimmer HelperCampground AttendantAir Purifier ServicerLight Fixture ServicerWeed Control InspectorBuilding Pressure WasherInsect Control InspectorDisease Control Inspector
The BLS uses Building Cleaning 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 Sweeping a flue by feel, snaking a brush through offset ducts, rigging a swing stage for a 30-story glass face, or scraping cured grout and mortar off post-construction floors are all irregular-geometry hand tasks with no two jobs alike — an 18 rather than 20 only because route scheduling, quoting square footage, and the flat-open-warehouse floor scrubbing at the margins are already going to software and ride-on autoscrubbers.
Hands-on in uncontrolled environments The work is ladders, roofs, crawlspaces, confined-space entries, pressure washers at 3,000+ psi, and respirators against creosote, mold, and bloodborne pathogens — uncontrolled environments in every sense, docked one point only because a slice of the category is interior mid-rise window work on solid footing.
No licence, no signature requirement No state licence gates the trade; a CSIA chimney sweep certification or an IICRC or OSHA 10 card helps you win bids and satisfy an insurer, but nothing stops an uncertified crew from underbidding you tomorrow — the 4 is those voluntary credentials, not legal protection.
Executes defined procedures on defined inputs Calls like whether a flue liner is cracked enough to condemn or which chemical to use on a stained façade are real but bounded by manufacturer specs, NFPA 211, and a supervisor's sign-off on the scope — you flag the problem, the customer or the GC decides, which is a 6 rather than discretionary authority.
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 (18/20 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 (8/20) 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 18 of this occupation's 55 points (33%).
Embodiment (19/20) 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 74/100 — SAFE.
Post-loss and post-construction work increasingly gates on documented clearance decisions — passing an independent third-party verification, calling whether a duct system is contaminated per NADCA ACR, deciding whether a fire-damaged flue is condemnable. If insurers require a named technician's contamination/clearance determination to release payment (mirroring IICRC S500/S520 use in water and mold claims), the role owns a consequential ambiguous call.
High-rise exterior work: OSHA 1910.27/ANSI I-14.1 window cleaning safety standards and building-owner rooftop anchor certification requirements push toward a designated competent person who signs the fall-protection plan per job. If NY-style facade/suspended-scaffold licensing (NYC DOB requires licensed riggers and certificate holders) is adopted more broadly, the signing competent person becomes a legally required human role.
Task-mix shift is real here but small: if floor-scrubbing robots (Tennant/Brain, Avidbots) and dispatch software absorb the flat-open-area and quoting tier, what remains is confined-space, roof, and contaminated work. The score is already 18 so headroom is minimal.
State-level licensing/certification mandates for specific specialty niches already exist in fragments and could spread: chimney sweeps (CSIA certification is voluntary; some jurisdictions like parts of NJ and CT tie hearth inspection sign-off to certified sweeps), NADCA ACR-certified air duct cleaning being written into commercial HVAC insurance and IAQ contracts, and trauma/biohazard cleanup requiring OSHA bloodborne-pathogen-trained supervisors plus state medical-waste transporter permits. If a state adopts a licensing act for biohazard remediation contractors (as several have proposed following mold-remediation licensing in FL, TX, LA), a named certified individual signs the clearance document and carries liability.
Narrow route only: crime-scene/decomposition cleanup and occupied-home chimney work where discretion and presence in a distressed household are the purchase. Buyers pay a premium for a specific vetted, background-checked, insured named crew. This does not generalize to pressure washing or duct work, where the buyer is a procurement line item.
The limit. Capability dimensions are already near maximum — the score of 55 is an institutional problem, not a technical one. Realistic ceiling is roughly 70, and only for the licensed/certified niches (biohazard, high-rise facade, certified chimney/duct). The pressure-washing and general specialty-cleaning bulk of this SOC has no plausible route to a licensing regime or a trust premium, and wage protection there depends on labor supply and OSHA enforcement, not on AI resistance.
| Los Angeles-Long Beach-Anaheim, CA | 930 | $44,590 +1% |
| Las Vegas-Henderson-North Las Vegas, NV | 630 | $44,280 +1% |
| Houston-Pasadena-The Woodlands, TX | 500 | $37,050 -16% |
| Philadelphia-Camden-Wilmington, PA-NJ-DE-MD | 460 | $50,660 +15% |
| San Francisco-Oakland-Fremont, CA | 390 | $45,430 +3% |
| Denver-Aurora-Centennial, CO | 320 | $47,050 +7% |
| San Diego-Chula Vista-Carlsbad, CA | 290 | $46,960 +7% |
| Portland-Vancouver-Hillsboro, OR-WA | 280 | $47,310 +7% |
| Boston-Cambridge-Newton, MA-NH | 130 | $70,570 +60% |
| Pittsburgh, PA | 130 | $51,370 +17% |
| Philadelphia-Camden-Wilmington, PA-NJ-DE-MD | 460 | $50,660 +15% |
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 55. 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.