EXPOSED
The paperwork half of this job — shift schedules, production reports, downtime logs, quality paperwork, performance write-ups, staffing forecasts — is squarely in reach of current AI plus MES/ERP automation, and many plants already push those tasks into software. The other half is standing on a line, reading a machine that sounds wrong, redeploying people when someone calls out at 5am, coaching a new operator's hands, and taking the hit when a lot ships out of spec; none of that is automatable today. The modal worker is a shop-floor supervisor of 10–30 operators, and the likely outcome is fewer supervisors covering wider spans with AI-generated schedules and reports.
Dipped in 2020, then grew past where it started.
Median pay $61,310 → $74,450 -2.9% 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
+1.2% 698,600 → 706,900 on the projections basis
Growing, and only partly exposed
The BLS expects +1.2% more of these jobs by 2034, and at 56/100 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.
~67,700 openings a year on average, including replacing people who leave.
Woods BossBeater BossTipple BossBreaker BossGauger ChiefShop ForemanPit SupervisorAcid SupervisorDock SupervisorDyer SupervisorLime SupervisorLine SupervisorMill SupervisorPipe SupervisorPond SupervisorShop SupervisorTurn SupervisorYard SupervisorBrine SupervisorCandy SupervisorConcrete ForemanDials SupervisorFence SupervisorGlaze Supervisor
Holding it up: embodiment . Weakest point: liability shield .
Mixed — a routine tier and a judgment tier Roughly half the shift is generating documents an MES already holds the data for — crew assignments, hourly production counts, scrap and downtime coding, OEE summaries, corrective-action forms — and that half is automatable now; the 12 rather than 8 comes from the tasks that require you to physically be at the press when it starts throwing flash, and to decide in thirty seconds whether to run it or lock it out.
Hands-on in uncontrolled environments You spend the shift walking the floor in PPE, putting hands on a jammed conveyor, standing over a new operator's setup to correct their grip on the fixture, and doing your own eyes-on verification of first-piece quality — the 16 rather than 20 reflects that a growing share of the job is done from a terminal in the shift office.
No licence, no signature requirement There is no licence to supervise production; a plant can promote a senior operator into your job on a Monday, and when an out-of-spec lot ships the exposure lands on the company and the quality system, not on your personal credential — forklift, lockout/tagout, or OSHA 30 cards are internal requirements, not statutory gatekeepers.
Meaningful discretion You make calls with real money and real bodies attached — stop the line or run to end of shift, put a marginal operator on the critical machine, judge whether a borderline dimension gets a deviation or gets scrapped — and you own the outcome at the Monday review; the 13 rather than 16 reflects that specs, SOPs, and a quality manager constrain most of those decisions before you reach them.
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 (12/20 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 (11/20) is whether buyers specifically pay for a person. Judgment and accountability (13/20) 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 28 of this occupation's 56 points (50%).
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.
No occupation passed every test: close enough to first-line supervisors of production and operating workers on skills and subject matter, at least 10 points more resistant, no big jump in training, no new licence, no pay cut, and not shrinking on its own. That happens for 223 of the 654 occupations here that aren't SAFE, and it is worth stating plainly rather than leaving the section off.
The usual reason is that exposure travels with the skill profile. The jobs most similar to yours tend to be exposed for the same reasons yours is, so the near neighbours don't clear the gap — and the ones that do are a different kind of work, not a transfer of what you already know. Read that as a limit of this method, not a verdict that you're stuck: it only compares whole occupations, and it cannot see specialisation, industry, or anything you'd bring that isn't in a federal skill survey.
Here is that claim on your own job rather than in the abstract. These are the three occupations closest to this one by skill and subject matter — the places the work would most naturally transfer — with what the register scores them:
That is the whole problem in three lines. The nearest work is not meaningfully safer, so there is no move here that trades a similar skill set for a better verdict. This is not us running out of ideas — it is what the neighbourhood looks like.
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 73/100 — SAFE.
Pure task-mix shift, no law needed: this job genuinely has two tiers. If MES/ERP plus LLM layers absorb scheduling, downtime coding, shift reports, quality paperwork and staffing forecasts, the residual role is the non-automatable tier — 5am redeployment under absenteeism, hands-on coaching of new operators, diagnosing a machine that sounds wrong, and owning the scrap call. The measured share of the day that AI cannot do at usable quality rises even as headcount falls. Watch for job postings that drop 'reporting' language and add 'coaching, troubleshooting, span of 40+'.
Automated-decision employment law reaching shop-floor discipline: NYC Local Law 144 and Illinois' AI Video Interview Act cover hiring; California's CRD automated-decision rules (effective Oct 2025) and proposed state bills reach broader employment decisions. A rule or arbitration line holding that write-ups, attendance terminations, or productivity-based discipline generated from MES/labor-tracking data require a named human supervisor as the accountable decision-maker (and expose that supervisor's judgment to just-cause review) makes the human signature legally load-bearing.
Extension of existing named-person sign-off regimes into general manufacturing: FDA 21 CFR 211.100/211.188 already requires a qualified person to review and sign batch production records, FSMA requires a named PCQI to sign corrective-action and verification records, and OSHA PSM (1910.119) requires certification of operating procedures by a qualified person. If OSHA's long-pending update to the PSM standard, or state food/cannabis/battery-plant rules, adds an explicit requirement that a named shift supervisor personally certify AI-generated deviation dispositions, lockout/tagout authorizations, or line-release decisions — and that the certification cannot be executed by software — the shield moves from a soft norm to a personal one.
Formalization of stop-work authority in the supervisor role: customer-driven quality standards (IATF 16949 clause 10.2 on customer complaints, AS9100 on escapes) and post-incident consent decrees increasingly name a specific on-shift authority for line stoppage and containment. If plant quality manuals or union safety agreements (e.g. UAW/USW health-and-safety language) designate the first-line supervisor as the sole holder of stop-work and product-hold authority, with that decision auditable to them personally, the consequential-call ownership hardens rather than diffusing into a central control room.
Customer- and insurer-driven presence requirements: aerospace, medical-device and pharma customers already write supervised-shift and named-contact clauses into supply agreements, and property/casualty insurers underwriting high-hazard lines can require documented supervisory coverage per shift as a condition of premium. If audit protocols or insurance riders specify a qualified human supervisor physically present per shift per line, buyers are effectively paying for the human. This is a narrow, sector-specific route and does not generalize to commodity contract manufacturing.
The limit. Embodiment is already near its practical maximum at 16 and has no upward route; if anything, lights-out cells and remote monitoring push it down. The binding constraint on this occupation is not per-worker defensibility but span of control: every lever above can fire and the count of supervisors still falls, because AI-generated schedules and reports let one defensible human cover 60 operators instead of 20. Levers here protect the role's content, not the headcount.
| Chicago-Naperville-Elgin, IL-IN | 20,970 | $76,500 +3% |
| New York-Newark-Jersey City, NY-NJ | 19,220 | $81,790 +10% |
| Los Angeles-Long Beach-Anaheim, CA | 17,650 | $74,740 +0% |
| Dallas-Fort Worth-Arlington, TX | 16,880 | $69,940 -6% |
| Houston-Pasadena-The Woodlands, TX | 15,270 | $75,330 +1% |
| Atlanta-Sandy Springs-Roswell, GA | 11,730 | $71,310 -4% |
| Detroit-Warren-Dearborn, MI | 10,380 | $75,490 +1% |
| Philadelphia-Camden-Wilmington, PA-NJ-DE-MD | 9,460 | $79,520 +7% |
| Bremerton-Silverdale-Port Orchard, WA | 500 | $115,340 +55% |
| Norwich-New London-Willimantic, CT | 1,000 | $105,690 +42% |
| Lexington Park, MD | 130 | $99,980 +34% |
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 56. 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.