← Risk register SOC 25-4012 · reviewed 2026-08-11

Curators

12,150 US workers · median $63,420/yr · Education

EXPOSED

A large share of a curator's week is text and screen work AI already does passably: cataloging records, wall labels, catalog essays, condition-report write-ups, grant narratives, and literature searches on provenance. What resists is the physical side — handling and installing objects, judging authenticity and condition in the room — plus the accountability for acquisition, deaccession, and repatriation calls that a board, donor, or source community will question for decades. There is no license behind the title, so the protection comes from institutional trust and scarcity of positions rather than from law.

10-year outlook: Curator headcount stays small and flat while AI absorbs the cataloging and label-writing load, so the surviving roles skew toward object expertise, acquisition ethics, and donor-facing accountability.

US employment, 2019–2025-5.7%
12,89012,150 workers

Nearly all of this fall was the 2020 shock. It has been climbing back since.

Median pay $54,570 → $63,420 -7.0% in real terms (nominal +16.2%, less ~25% US inflation over the period)

The job count is not the verdict

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

+7%

Percentage only. The projection counts a different population from the 12,150 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 +7% more of these jobs by 2034, and at 49/100 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.

~1,800 openings a year on average, including replacing people who leave.

One email if this score changes. Watch as many occupations as you like from the same address — no account, and nothing is sent on a schedule, only when a verdict actually moves.

Also known as — 24 job titles this covers

Titles reported by people doing this work, from the US Department of Labor's O*NET survey. If your job title is here, this page is about your work even though the name doesn't match.

CuratorPreparatorArt CuratorArt HandlerNumismatistPhilatelistData CuratorCoin CollectorMuseum CuratorMuseum ManagerContent CuratorDigital CuratorField CollectorMuseum EducatorOld Coin DealerStamp CollectorExhibits CuratorMetadata CuratorMuseum RegistrarEducation CuratorHerbarium CuratorMuseum SpecialistMuseum CoordinatorCollections Curator

Score — 49/100 resistance

Holding it up: judgment & accountability (13/20). Weakest point: liability shield (3/20).

Five dimensions, 0–20 each, summed. Higher means more protected. The arithmetic is shown so you can check it: 11 + 11 + 3 + 11 + 13 = 49. · Scored 2026-08-11, and re-examined when evidence accumulates rather than on a schedule.

Task resistance 11/20

Mixed — a routine tier and a judgment tier The write-heavy half of the job — TMS/PastPerfect data entry, label copy, exhibition didactics, loan-agreement paperwork, grant reporting to IMLS or NEH — is now draftable by machine, but selecting what enters an exhibition, negotiating loans with a peer institution's registrar, and standing in front of an unframed canvas deciding whether the craquelure reads as period keep this at 11 rather than in single digits.

Embodiment 11/20

Some physical or field component Curators are in storage vaults and on gallery floors more than the title suggests — checking RH and lux readings, condition-reporting incoming crates, directing mount-making and hang heights, doing site visits to a collector's damp basement — but preparators and conservators do the actual heavy handling and interventive treatment, which caps it at 11 instead of the 15+ of a field conservator.

Liability shield 3/20

No licence, no signature requirement There is no state license to curate; AAM accreditation and AAMD guidelines bind the institution, not the individual, and a curator who mis-attributes a work or accepts a piece with a broken provenance chain faces reputational and employment consequences while the museum's directors and insurers absorb the legal exposure — hence 3, not 0, since the field polices credentials informally through a small, long-memoried professional network.

Trust premium 11/20

Some relationship component Donors give collections to a person they have cultivated over years, source communities under NAGPRA consult with the individual who returned their calls, and scholars lend on the strength of a named curator's reputation — but a large share of curatorial output reaches the public anonymously as an unsigned label or an object record, which is why this sits at 11 rather than the 16+ of a private dealer.

Judgment & accountability 13/20

Meaningful discretion Acquisition, deaccession, attribution, and repatriation decisions are made on incomplete provenance in the face of competing claims, and they are permanent — a wrongly deaccessioned object does not come back, and a NAGPRA determination or a Nazi-era restitution call will be re-litigated by successors and press for decades — with the ceiling set only by the fact that boards and directors formally sign off.

Confidence: medium · reviewed 2026-08-11 · how scoring works

What this job involves — and which parts are yours

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.

AI already does these at usable quality

These still need a person

Active moats on the surviving side: judgment, trust, embodiment

How to future-proof this job

Training paths for your skill gaps: Coursera — teaching and instructional design, audit free free to audit · Learning How to Learn — the most-taken course on Coursera, and free free to audit · Coursera — project coordination and cross-team delivery free to audit · Coursera — communication and interpersonal skills free to audit · Coursera — customer service and client-facing skill courses free to audit · Coursera — negotiation, influence and persuasion courses free to audit

All 35 skills ranked by how many jobs they open →

Where this experience transfers — occupations you could move toward

Computed from U.S. Dept. of Labor O*NET skill and knowledge profiles: high overlap with what you already do, a materially higher resistance score, no large jump in required training, and no licence you would have to start a new pipeline to get. Targets that pay meaningfully less, that are themselves COOKED, or whose own headcount is falling are excluded — a move into a shrinking trade is not an escape.

Secondary School Teachers, Except Special and Career/Technical Education SAFE · 69/100 · you already have ~66% of the skill profile

Skills to close: Instructing, Learning Strategies, Coordination, Social Perceptiveness

Clergy SAFE · 69/100 · you already have ~66% of the skill profile

Skills to close: Service Orientation, Instructing, Persuasion, Social Perceptiveness

Middle School Teachers, Except Special and Career/Technical Education SAFE · 71/100 · you already have ~63% of the skill profile

Skills to close: Instructing, Learning Strategies, Coordination, Social Perceptiveness

What would move this back up — beyond any one person

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 65/100, still EXPOSED.

5 specific changes that would raise this score
  • already happening liability shield +4

    NAGPRA's revised regulations (43 CFR 10, effective January 2024) already require a named institutional official to certify inventories, obtain documented tribal consent before exhibiting or researching cultural items, and expose institutions to civil penalties. If federal agencies begin assessing penalties against named individuals, or if grant conditions (IMLS, NEH) require a named curator of record to attest to compliance, the sign-off becomes personal rather than institutional.

  • already happening judgment accountability +3

    Restitution decisions under the HEAR Act and the emerging African-collections repatriation frameworks (Benin Bronzes precedents, Smithsonian's 2022 ethical returns policy) put a named curator's recommendation in front of a board and, increasingly, a press cycle and a claimant's counsel. If ethical-returns policies spread to state and university museums with a documented curatorial recommendation of record, the consequential-call share of the job rises.

  • plausible liability shield +3

    Export/import controls already require a human attestation: CITES permits for ivory, tortoiseshell and feather in collections; State Department cultural-property MOU certifications; UK/EU export licence applications. If customs or insurers begin requiring that the condition and authenticity statement on loan and export paperwork be signed by a named individual who is personally attestable — as fine-art underwriters (AXA XL, Hiscox) already push toward for high-value loans — the countersignature attaches to the curator.

  • plausible trust premium +3

    A wave of AI-fabricated provenance documents entering the market (plausible given how much of provenance research is now text-matching) would make buyer-side demand shift toward named human attribution. Watch for auction houses and AAMD member museums adopting a rule that provenance chains must be certified by a named scholar who inspected physical documents, analogous to the Art Loss Register's human-review requirement.

  • plausible task resistance +3

    Genuine two-tier structure: label writing, catalog data entry, condition-report prose and grant narrative drafting are the routine tier and are being automated first. If institutions absorb that automation without cutting headcount — the pattern in small museums where the curator is also registrar and installer — the residual job is authenticity judgment in the room, acquisition argument, and community consultation, and measured resistance rises without any capability change.

The limit. No licensure exists for the title and no professional body has proposed one; AAM accreditation certifies institutions, not people, so the liability_shield route caps well short of the medical or engineering pattern. Trust premium is also constrained by field economics: 12,150 positions concentrated in budget-constrained institutions means willingness-to-pay for a human curator is set by boards and donors, not by a broad buyer market.

These are conditions, not forecasts — what would have to happen, not what will. Specific rules, cases and bills are named so you can go and check whether they exist and where they stand; verify before relying on any of them. Nothing here is legal or financial advice.

Where this work is, and what it pays there

BLS metro figures for 80 areas. The verdict above does not change by city — the rubric judges what the work involves, not where it happens — but pay and headcount do, and the national median hides a very wide range.

Most of these jobs

New York-Newark-Jersey City, NY-NJ 970 $83,110 +31%
Los Angeles-Long Beach-Anaheim, CA 390 $77,160 +22%
Chicago-Naperville-Elgin, IL-IN 290 $63,060 -1%
Philadelphia-Camden-Wilmington, PA-NJ-DE-MD 290 $63,860 +1%
Washington-Arlington-Alexandria, DC-VA-MD-WV 270 $87,400 +38%
San Francisco-Oakland-Fremont, CA 240 $90,800 +43%
Boston-Cambridge-Newton, MA-NH 220 $78,720 +24%
Denver-Aurora-Centennial, CO 180 $77,700 +23%

Best paid

New Haven, CT 70 $102,110 +61%
Dayton-Kettering-Beavercreek, OH 40 $94,740 +49%
San Francisco-Oakland-Fremont, CA 240 $90,800 +43%

Percentages are against this occupation's national median of $63,420. Counts are jobs in that metro, not vacancies. Metros where the BLS suppressed the cell are absent rather than shown as zero.

Who is actually doing this — nobody, on the record

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 49. 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.

Read that as a gap in the reporting we can see, not proof of absence — the dispatch runs on English-language feeds and misses plenty. If you know of a case, tell us, or add a field report from inside the job.

Quick take — do you do this job?

Has AI actually changed your work? One tap, anonymous, and the running tally is public. Nothing else is asked of you.

Self-reported and unverified — a sentiment signal, not a survey. One response per person per occupation; you can change your answer.

Field reports — what people say has changed

No field reports yet. A written account takes a paragraph rather than a tap, goes to an editor before it appears, and is the one thing on this page the rubric cannot produce on its own.

File a field report

Concrete beats general: a tool that arrived, a task that moved, a headcount decision you watched happen. Don't include anything that identifies you or your employer if that would put you at risk.

Watch this verdict
Kept current

Rather than check back: get the digest and we'll tell you what changed — or watch a single occupation from its own page.