EXPOSED
Nothing about running a paver, roller, or tamper is text or screen work, so language AI touches almost none of it — the real threat is machine autonomy, and paving is the most automatable niche in construction because the work is linear, repetitive, and GPS-guidable. Automated screed control, intelligent compaction with real-time density feedback, and stringless 3D paving already reduce headcount per crew and shift the job toward monitoring. What survives is judging mix temperature and workability, reading a subgrade that isn't behaving, hand-tamping edges and joints, and working safely inches from traffic and other crew.
Part 2020 shock, part continued decline in the years since.
Median pay $40,130 → $53,340 +6.3% 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
+3.2% 47,000 → 48,600 on the projections basis
Growing, and only partly exposed
The BLS expects +3.2% more of these jobs by 2034, and at 53/100 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.
~4,000 openings a year on average, including replacing people who leave.
Crew MemberForm GraderForm TamperAsphalt PaverAsphalt RakerPaver OperatorLoader OperatorPaving OperatorRoller OperatorScreed OperatorBlack Top RollerMachine OperatorAsphalt ScreedmanMud Jack OperatorPaving Crew MemberRoad Mixer OperatorForm Tamper OperatorRoad Packer OperatorRoad Roller OperatorCurb Machine OperatorSteam Roller OperatorAsphalt Paver OperatorAsphalt Plant OperatorHeater Planer Operator
Holding it up: embodiment . Weakest point: liability shield .
Tasks largely resist digitisation Screed operators still set head-of-material by eye, call for a slower truck exchange when the mat starts tearing, and drop the tamper on a manhole collar or driveway apron that no 3D model captured — but the mainline pass on a mile of interstate, the part that fills most of the shift, is exactly the linear GPS-guidable work that stringless paving and intelligent compaction already run with less human input, which keeps this at 15 rather than the 18 a framer or ironworker earns.
Hands-on in uncontrolled environments An 18 reflects standing on a machine deck over 300°F asphalt in July, shoveling and raking joints by hand, walking a vibratory plate around utility cuts, and doing all of it inside a live work zone where the environment changes every hour with weather, traffic, and whatever the subgrade does under load — no part of this is reproducible in a controlled cell.
No licence, no signature requirement There is no state licence to operate a paver or roller; a CDL to move the equipment and an OSHA 10/30 card are the practical ceiling, and when a mat fails density testing the liability lands on the contractor and the inspector's report, not on the operator's name — 4 rather than 0 only because CDL and flagger certification are real gates on hiring.
Meaningful discretion A 10 covers calls that are genuinely yours and genuinely consequential — how many roller passes before the mat cools out of the compaction window, whether the mix arrived too stiff to place, when to stop paving because the base is pumping — but they are made against QC specs, a lift thickness on the plan, and a foreman standing right there, so you are exercising discretion inside a spec rather than owning the ambiguous call alone.
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 (4/20) is whether the law requires a licensed human to sign. Trust premium (6/20) is whether buyers specifically pay for a person. Judgment and accountability (10/20) 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 20 of this occupation's 53 points (38%).
Embodiment (18/20) 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 64/100, still EXPOSED.
Task-mix shift is genuinely bimodal here: if intelligent compaction and stringless 3D screed control absorb the mainline mat, the residual job is the non-linear tier — tie-ins, cul-de-sacs, utility patches, driveway aprons, manhole and gutter joints, hand-tamping behind the machine, and reading a subgrade that pumps. Watch for state DOT specs (Minnesota DOT, Georgia DOT IC pilot specs) that mandate IC on mainline while still requiring hand methods at structures; that formalizes the split rather than eliminating the human tier.
Paving has no personal license, but two real mechanisms could create a signer. First, prevailing-wage and apprenticeship provisions: federal-aid and IIJA-funded projects already carry Davis-Bacon classification requirements, and some state DOTs require a certified roller/screed operator (e.g., asphalt paving certification programs run through state asphalt pavement associations and NHI training) to be present for acceptance of the mat. Second, if a state DOT makes density/smoothness acceptance contingent on a named certified operator's signed daily paving log — the way concrete testing requires an ACI-certified technician — that puts a specific human's credential on the pay item.
If warranty-based pavement contracts spread (Michigan DOT and several European models use multi-year pavement warranties), the contractor bears smoothness and density risk for years, and the field call to stop paving on a marginal mix temperature, a rainy window, or a soft subgrade becomes a consequential, attributable decision rather than a foreman's instruction. Watch for warranty clauses paired with named-operator daily logs.
Already near ceiling at 18. The only route up is enforceable rules keeping a human physically on the mat: an FHWA or state work-zone rule barring unmanned equipment operation within an active traffic-exposed work zone without an on-board operator, similar to how autonomous haul trucks are confined to closed mine sites. If OSHA or a state DOT wrote 'no unoccupied self-propelled paving equipment on open roadway', the residual environment stays unpredictable by definition.
The limit. Trust premium has no realistic route: pavement buyers are DOTs and developers who purchase a specification (density, thickness, IRI smoothness), not a craftsman, and nobody pays extra for a human-driven roller. Even with every lever above, the score lands in the mid-60s at best — the ceiling is set by the fact that mainline paving is the most geometrically automatable work in construction, and the levers protect the shrinking irregular remainder rather than the headcount.
| New York-Newark-Jersey City, NY-NJ | 3,440 | $62,780 +18% |
| Washington-Arlington-Alexandria, DC-VA-MD-WV | 980 | $58,800 +10% |
| Dallas-Fort Worth-Arlington, TX | 940 | $49,740 -7% |
| Houston-Pasadena-The Woodlands, TX | 870 | $46,790 -12% |
| Atlanta-Sandy Springs-Roswell, GA | 770 | $46,120 -14% |
| Chicago-Naperville-Elgin, IL-IN | 740 | $80,340 +51% |
| Detroit-Warren-Dearborn, MI | 580 | $51,310 -4% |
| Miami-Fort Lauderdale-West Palm Beach, FL | 570 | $46,810 -12% |
| Riverside-San Bernardino-Ontario, CA | 310 | $126,640 +137% |
| San Francisco-Oakland-Fremont, CA | 260 | $104,530 +96% |
| Seattle-Tacoma-Bellevue, WA | 270 | $96,590 +81% |
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 53. 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.