EXPOSED
This is a residual bucket — everything from program managers and operations managers to compliance and facilities leads — so the modal worker is a mid-level manager who supervises 5–20 people, owns a budget line, and reports upward. The reporting layer of that job (status decks, variance write-ups, meeting summaries, scheduling, policy drafting, KPI dashboards) is already largely automatable, while hiring, firing, performance calls, cross-team negotiation, and owning a bad outcome are not. The real risk is span-of-control compression: AI reporting tools let one manager cover more people, so headcount thins rather than the role disappearing.
Headcount grew steadily across the period.
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
+4.5%
Percentage only. The projection counts a different population from the 622,190 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 +4.5% more of these jobs by 2034, and at 46/100 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.
~106,700 openings a year on average, including replacing people who leave.
ConsulPublisherRegistrarChief ClerkGym ManagerTeam LeaderWatermasterBranch ChiefCamp ManagerManufacturerPool ManagerRisk ManagerSite ManagerAudit ManagerCamp DirectorGroup ManagerHouse ManagerShift ManagerTraffic ChiefChancery ClerkClerk of CourtDebris MonitorDivision ChiefEthics Manager
The BLS uses Managers, 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: judgment & accountability . Weakest point: liability shield .
Mixed — a routine tier and a judgment tier The recurring artifacts you produce — weekly status decks, budget variance narratives, headcount forecasts, SOP redlines, action-item tracking out of meetings — are exactly what LLMs draft competently from your own systems of record, but the parts that put this at 11 rather than 5 are the ones with no source document: telling a strong performer they didn't get the promotion, deciding which of two departments eats the shortfall, and holding a vendor to a commitment that was never written down.
Some physical or field component A 6 reflects that most of your week is Teams, spreadsheets, and a conference room, with a real but bounded physical tail — walking the floor or site, being present for an investigation or a termination, showing up when a facility, shipment, or event goes sideways — none of which requires you to handle equipment or work in conditions a screen couldn't reach.
No licence, no signature requirement Nothing in this bucket requires a licence to hold the title; PMP, SHRM-CP, or CFM help you get hired but no board can pull them in a way that stops someone else from doing your job, and when something goes wrong the exposure lands on the entity or on an officer who signs, not on you personally.
Exists to be accountable for ambiguous calls A 15 is about ownership of calls with no correct answer and no one above you to absorb them: which of five underfunded priorities gets cut, whether a performance problem is coachable or terminal, when to escalate a compliance concern that will embarrass a colleague — and if the project misses, the postmortem names you, not the tooling that produced your forecast.
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 (11/20 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 (2/20) 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 (15/20) 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 29 of this occupation's 46 points (63%).
Embodiment (6/20) 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 two-tiered here: if reporting/deck/variance/scheduling work is fully absorbed by AI, the residual day is hiring, firing, performance calibration, cross-team negotiation and escalation handling — none currently automatable. Watch for job postings in this bucket dropping 'reporting' language and adding 'people leadership' as the sole core duty.
If AI systems make more of the recommendations, the manager's remaining role is override-and-own: documented deviation from an AI recommendation, and bearing the outcome. Watch for internal governance policies (already appearing in bank and insurer model-risk frameworks) that require a manager to log a written rationale whenever they overrule a model.
Named-human-accountability rules attaching to management decisions: EU AI Act Art. 14 human-oversight duties for high-risk HR/employment systems require an identified natural person able to override; NYC Local Law 144 and Illinois AI Video Interview Act already put a named employer representative behind automated hiring tools. If audit regimes require a named manager to sign off on each AI-assisted hiring, promotion, or termination decision, the sign-off becomes a job function rather than a formality.
Sector-specific personal accountability regimes extending to operations/compliance leads — e.g. the UK Senior Managers & Certification Regime (individual FCA-registered accountability for a defined function), or OSHA/EPA-style named responsible-person designations for facilities managers. Where a statute names an individual rather than the firm, span-of-control compression stalls because one person cannot hold unlimited designated functions.
Narrow route only: union contracts requiring that discipline, grievance and scheduling decisions be made by an identified human supervisor with authority to deviate — e.g. Teamsters' 2023 UPS contract language on technology and driver-facing monitoring, and existing US port and rail agreements restricting automated dispatch. This raises trust premium for the represented-workforce slice of this bucket, not for corporate program managers.
The limit. The binding constraint is span-of-control compression, which none of these levers fully addresses: a liability shield or override duty can survive while the number of managers needed halves, because one named signer can cover a larger team. Trust premium is also structurally weak — the buyer of a mid-level manager's output is an internal executive optimising cost, not an external customer who can be charged extra for a human. Realistic ceiling around 60.
| Washington-Arlington-Alexandria, DC-VA-MD-WV | 42,010 | $170,980 +20% |
| New York-Newark-Jersey City, NY-NJ | 34,400 | $169,800 +20% |
| Los Angeles-Long Beach-Anaheim, CA | 31,210 | $168,160 +19% |
| Chicago-Naperville-Elgin, IL-IN | 26,170 | $145,370 +2% |
| Atlanta-Sandy Springs-Roswell, GA | 24,490 | $130,340 -8% |
| San Francisco-Oakland-Fremont, CA | 21,550 | $210,080 +48% |
| Dallas-Fort Worth-Arlington, TX | 19,890 | $139,810 -1% |
| Houston-Pasadena-The Woodlands, TX | 14,930 | $144,820 +2% |
| San Jose-Sunnyvale-Santa Clara, CA | 13,740 | $217,280 +53% |
| San Francisco-Oakland-Fremont, CA | 21,550 | $210,080 +48% |
| Boston-Cambridge-Newton, MA-NH | 9,120 | $184,250 +30% |
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 46. 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.