← Risk register SOC 43-6014 · reviewed 2026-08-11

Secretaries and Administrative Assistants, Except Legal, Medical, and Executive

1,706,790 US workers · median $47,540/yr · Office

COOKED

The core of this job — scheduling meetings, drafting and formatting correspondence, transcribing notes, maintaining filing systems, screening and routing calls and email, entering data into spreadsheets and CRMs — is precisely what current AI assistants and calendar/workflow automation already do at usable quality. What resists is the physical, in-office layer: receiving visitors and deliveries, handling paper and equipment, walking a document to a person, holding the informal knowledge of how a particular office actually runs. No licensure, no signature requirement, and buyers are not paying for a relationship with the admin, so the only real moat is on-site presence and office-specific judgment.

10-year outlook: Expect continued steady shrinkage of headcount through the 2030s as AI absorbs the correspondence, scheduling, and data-entry core, with surviving roles consolidating into fewer, broader office-operations and coordination jobs that require being physically present.

US employment, 2019–2025-16.3%
2,038,3401,706,790 workers

Part 2020 shock, part continued decline in the years since.

Median pay $37,690 → $47,540 +0.9% in real terms (nominal +26.1%, 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

-1.6% 1,944,000 → 1,913,200 on the projections basis

Exposed, and shrinking

Both signals point the same way: the tasks are largely automatable and the BLS projects -1.6% by 2034. This is the case where the score and the forecast agree, and it is the one worth taking seriously.

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.

~202,800 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.

PoliceSecretaryClerk SecretaryStaff AssistantAlumni SecretaryEscrow AssistantEscrow SecretaryOffice AssistantOffice SecretarySchool SecretaryProgram AssistantClerical SecretaryPersonal AssistantPersonal SecretaryAdministrative ClerkDepartment SecretaryOffice AdministratorOperations AssistantAppointment SchedulerAdministrative LiaisonFront Desk CoordinatorScheduling CoordinatorAdministrative AssociateAdministrative Technician

Score — 22/100 resistance

Holding it up: trust premium (7/20). Weakest point: liability shield (0/20).

Five dimensions, 0–20 each, summed. Higher means more protected. The arithmetic is shown so you can check it: 4 + 6 + 0 + 7 + 5 = 22. · Scored 2026-08-11, and re-examined when evidence accumulates rather than on a schedule.

Task resistance 4/20

Core tasks are already automatable Calendar coordination, email triage, meeting-minute transcription, mail merges, travel booking, expense-report entry and spreadsheet upkeep are each already shipped features in Outlook/Google Workspace plus an LLM, which puts this at 4 rather than 10 — the resistant residue is errand-running and improvisation when the printer jams, not any of the named duties.

Embodiment 6/20

Some physical or field component A 6 reflects that the physical work is real but indoor and low-variance: greeting visitors, signing for couriers, restocking supply cabinets, badging in guests, wrestling the copier and moving paper files — controlled office floor, no ladders, weather, or patient contact, so nowhere near the 13+ band.

Liability shield 0/20

No licence, no signature requirement There is no credential gate at all: no state licence, no mandatory certification, and the CAP/MOS credentials are resume decoration rather than a legal precondition — memos and calendar invites carry the boss's authority, and errors are absorbed by the supervisor, so 0 is literal, not a slight.

Trust premium 7/20

Some relationship component 7 credits the accumulated office-specific relationships — knowing which VP will actually take a call, which vendor to chase, how a department's approval chain really works — but the position is filled through temp agencies and turned over without clients noticing, so the relationship is convenience, not the product being bought.

Judgment & accountability 5/20

Executes defined procedures on defined inputs Most decisions run on standing rules — whose calendar wins a conflict, which template to use, what gets escalated — and the genuine discretion (choosing not to interrupt, resolving a double-booking, deciding a caller matters) is real-time and reversible within the day, which is a 5 rather than the 10+ of someone owning irreversible commitments.

Confidence: high · 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: physical-presence

How to future-proof this job

Training paths for your skill gaps: Khan Academy — mathematics, arithmetic through calculus free · Coursera — critical thinking and logic, audit free free to audit · 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 · MIT OpenCourseWare — circuits, systems and diagnostic method, free and ungated free · Apprenticeship.gov — industrial maintenance and millwright programs paid to train · OSHA Outreach Training — the 10- and 30-hour cards most employers ask for low · MIT OpenCourseWare — systems analysis and engineering free · MIT OpenCourseWare — problem-solving and analytical method courses free

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.

Tax Examiners and Collectors, and Revenue Agents EXPOSED · 34/100 · you already have ~64% of the skill profile

Skills to close: Mathematics, Critical Thinking, Instructing, Learning Strategies

Court Reporters and Simultaneous Captioners EXPOSED · 34/100 · you already have ~64% of the skill profile

Skills to close: Troubleshooting, Equipment Maintenance, Operation and Control

Private Detectives and Investigators EXPOSED · 56/100 · you already have ~62% of the skill profile

Skills to close: Critical Thinking, Systems Analysis, Complex Problem Solving, Operation and Control

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 35/100 — EXPOSED.

4 specific changes that would raise this score
  • already happening embodiment +4

    Return-to-office mandates plus consolidation of admin roles into single on-site 'office coordinator' posts that absorb facilities, mail/package intake, badge and visitor control, equipment troubleshooting, and event setup — already visible in JPMorgan/Amazon 5-day RTO staffing and in postings retitled Office Coordinator/Workplace Experience Coordinator. The physical, unpredictable-environment share of the day rises even as keyboard work falls.

  • plausible task resistance +3

    Genuine two-tier split: if the calendar/email/data-entry tier is fully automated, what remains is exception handling — vendor disputes, reconciling conflicting executive priorities, catching the scheduling conflict the tool didn't know about, onboarding logistics for a new hire. Watch for job postings where 'AI tool oversight' and 'verify AI-drafted correspondence before send' appear as listed duties.

  • plausible liability shield +3

    Narrow but real: admins who serve as notaries public, or as designated records custodians under state public-records/retention statutes (common in school districts, municipal offices, clerk's offices), carry a personal signature and statutory duty. If employers formalize a named human records custodian for AI-generated document retention — the direction of state RIM rules and of e-discovery authenticity fights over AI-produced records under FRE 901(b)(9) — a signature requirement attaches to the role.

  • plausible judgment accountability +3

    Designation as the human in the loop for confidential-information routing and data-handling under state privacy laws or an employer's AI-use policy — e.g., the person who decides what goes into a chatbot, who gets access to a personnel file, which caller is a social-engineering attempt. Insurer cyber policies increasingly require a named employee for wire-transfer and vendor-change verification after BEC losses.

The limit. Trust premium has no plausible route: buyers of this role are employers, not clients, and no one purchases a service specifically because a non-executive admin performed it. Even with every lever above firing, the occupation is capped in the 35-40 range — the levers convert an admin into an office coordinator or records custodian, which is a different job with fewer seats, not a defended version of 1.7M existing ones.

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 393 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 122,680 $51,440 +8%
Los Angeles-Long Beach-Anaheim, CA 58,800 $53,550 +13%
Chicago-Naperville-Elgin, IL-IN 50,190 $50,440 +6%
Dallas-Fort Worth-Arlington, TX 37,550 $47,450 +0%
Washington-Arlington-Alexandria, DC-VA-MD-WV 37,010 $53,660 +13%
Houston-Pasadena-The Woodlands, TX 33,710 $46,750 -2%
Miami-Fort Lauderdale-West Palm Beach, FL 33,380 $47,450 +0%
Philadelphia-Camden-Wilmington, PA-NJ-DE-MD 32,850 $48,850 +3%

Best paid

New Haven, CT 4,780 $63,800 +34%
San Jose-Sunnyvale-Santa Clara, CA 7,970 $62,940 +32%
San Francisco-Oakland-Fremont, CA 20,270 $62,110 +31%

Percentages are against this occupation's national median of $47,540. 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

The score above is about what the work exposes. This is reporting about real deployments in this occupation — the difference between "could be automated" and "somebody automated it."

1 of 1 reported case, with sources

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.