EXPOSED
This catch-all bucket holds instructional support staff, learning-center and tutoring coordinators, curriculum and media assistants, and program aides; the modal worker splits time between screen work (building lesson materials, cataloging resources, scheduling, tracking student progress, answering reference and referral questions) and in-person contact with students on a campus or in a library. The screen half is exactly what generative AI does cheaply — worksheet generation, reading-level adaptation, resource summaries, routine research help. The in-person half — sitting with a struggling student, running a study skills group, keeping a program physically operating — holds, but it carries no license and no signature requirement, so the only real moat is presence and relationship.
Most of this decline happened after 2021 — it is not the pandemic dip.
Median pay $40,310 → $50,890 +1.0% 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
+1.5% 132,000 → 134,000 on the projections basis
Exposed, but growing
AI can already do a lot of these tasks, and the BLS still expects +1.5% more of these jobs by 2034. Demand for the output is growing faster than the work is being automated away — the mechanism BLS gives for software developers, and the combination people most often misread as an error.
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.
~12,500 openings a year on average, including replacing people who leave.
Teacher AideCraft DemonstratorAptitude Test GraderEducation TechnicianEncyclopedia Research WorkerSatellite Instruction FacilitatorScholastic Aptitude Test (SAT) GraderIndividualized Education Plan (IEP) AideGeneral Educational Development (GED) ExaminerGED Examiner (General Educational Development Examiner)
The BLS uses Educational Instruction and Library Workers, 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: trust premium . Weakest point: liability shield .
Mixed — a routine tier and a judgment tier Roughly half the workday — drafting worksheets, adapting texts to reading level, writing resource guides, pulling citations for a student's paper, logging attendance and progress notes — is text-in/text-out work a model does in seconds, while the other half (sitting beside a student who won't start their essay, running a small-group study skills session, setting up a maker space or testing room) doesn't compress, which is why this lands at 9 rather than in the resistant teens.
Some physical or field component You are on a campus or in a library building: shelving and processing materials, moving AV carts and laptops, staffing a tutoring desk, proctoring a testing room, supervising a group during an after-school block — physical but in controlled, climate-conditioned, predictable spaces, which puts it at 10 rather than the 15+ of a field technician or trades instructor.
No licence, no signature requirement Most of these positions require a bachelor's or less with no state credential, no teaching license, and no professional certification a state board could pull — the certified teacher of record or the head librarian signs off on grades, IEP paperwork, and collection decisions, so the 3 reflects a real absence of statutory gatekeeping, not merely weak protection.
Meaningful discretion You decide which intervention a struggling reader gets next, when to escalate a behaviour or welfare concern, and how to reallocate a session when three students show up needing different things — real discretion inside a curriculum, a district handbook and a mandated-reporting chain that specifies the escalation, which is why it sits at 8 and not in the mid-teens where the final call is yours to own.
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 (9/20 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 (3/20) 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 (8/20) 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 22 of this occupation's 41 points (54%).
Embodiment (10/20) 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 59/100, still EXPOSED.
Task-mix shift: if worksheet generation, reading-level adaptation, cataloging and routine reference are absorbed by district-licensed AI tools (Khanmigo-type deployments, ILS AI discovery layers), the residual day becomes small-group intervention, behavior de-escalation, attendance/truancy follow-up and MTSS/RTI progress monitoring with a live child — work current models cannot do. Watch for job postings that drop 'creates instructional materials' and add 'delivers Tier 2 intervention'.
Being formally seated on MTSS/RTI or threat-assessment teams where the role owns the recommendation to escalate a student — several states now mandate multidisciplinary behavioral threat assessment teams (e.g., Virginia, Texas SB 11, Florida) and support staff are named members. Ownership of an escalation call under ambiguity is a real accountability transfer, not a title change.
Mandatory-reporter and child-supervision duties being formalized into the role description with personal legal exposure (state child-abuse reporting statutes already name school employees), plus state student-data-privacy rules (Illinois SOPPA, NY Ed Law 2-d) that require a designated human to approve any AI tool touching student records.
State or district rules making a named, certified human the required signer on IEP/504 service logs and progress data used in eligibility decisions — i.e., barring AI-generated progress notes from being entered without a credentialed staff attestation. IDEA documentation audits and state paraprofessional certification (ESSA Title I para requirements) are the existing hooks; several states are drafting AI-use rules for special-education documentation.
State legislation requiring a human educator of record and prohibiting AI as the sole provider of instruction or tutoring hours — bills of this shape have appeared in several statehouses and in teacher-union contract bargaining (UTLA, Chicago Teachers Union AI language). If high-dosage tutoring funding is conditioned on human-delivered contact minutes, buyer preference becomes a purchasing rule.
Parent-facing demand: private learning centers and library programs explicitly marketing 'no AI tutor, a person sits with your child', and public library systems adopting ALA-aligned policies that reference questions receive human review. Observable in center advertising and in library AI policy statements.
The limit. This is a catch-all SOC with no license, no scope-of-practice statute and no protected title, so the liability_shield route is capped: attestation duties can be added, but personal malpractice liability of the kind physicians or engineers carry is not on any visible path. The realistic ceiling is mid-50s, and it depends almost entirely on the job being redefined around in-person intervention and escalation rather than material production — a redefinition that in practice often comes with headcount cuts to the material-production half.
| Los Angeles-Long Beach-Anaheim, CA | 15,120 | $63,020 +24% |
| San Francisco-Oakland-Fremont, CA | 5,740 | $75,420 +48% |
| Atlanta-Sandy Springs-Roswell, GA | 3,340 | $25,000 -51% |
| San Diego-Chula Vista-Carlsbad, CA | 3,170 | $62,230 +22% |
| San Jose-Sunnyvale-Santa Clara, CA | 2,710 | $79,270 +56% |
| Riverside-San Bernardino-Ontario, CA | 2,590 | $66,920 +31% |
| Dallas-Fort Worth-Arlington, TX | 2,240 | $46,380 -9% |
| Portland-Vancouver-Hillsboro, OR-WA | 2,110 | $49,540 -3% |
| Grand Junction, CO | 40 | $81,730 +61% |
| Kahului-Wailuku, HI | 80 | $80,300 +58% |
| San Jose-Sunnyvale-Santa Clara, CA | 2,710 | $79,270 +56% |
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 41. 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.