SAFE
Middle school teaching is 80% classroom management, adolescent motivation, and physical supervision of 25 twelve-year-olds — none of which AI touches. Lesson plans, worksheets, quiz banks, rubric-based grading, and parent-email drafting are already being generated by AI, which will absorb prep hours rather than jobs. State licensure, mandatory-reporter duties, and in-loco-parentis custody of minors make an adult body in the room legally non-optional.
Roughly flat across the period, with year-to-year wobble.
Median pay $59,660 → $64,370 -13.7% 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
-2% 633,700 → 621,300 on the projections basis
Hard to automate, but shrinking anyway
The work resists current AI, yet the BLS projects -2% by 2034. Whatever is shrinking this occupation, the evidence does not point to automation — demand, demographics, offshoring and industry decline all shrink jobs that no machine could do.
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.
~40,500 openings a year on average, including replacing people who leave.
CoachTeacherEducatorArt TeacherArt EducatorBand TeacherHome TeacherBand DirectorMusic TeacherArt InstructorHealth TeacherMusic EducatorAlgebra TeacherEnglish TeacherHistory TeacherReading TeacherScience TeacherSpanish TeacherComputer TeacherBilingual TeacherClassroom TeacherOrchestra TeacherHumanities TeacherMontessori Teacher
Holding it up: trust premium . Weakest point: liability shield .
Mixed — a routine tier and a judgment tier The gradeable artifacts — unit plans aligned to state standards, differentiated worksheets, quiz banks, rubric scoring of five-paragraph essays — are already machine-producible, which caps this below 14, but the daily core of redirecting an off-task 13-year-old, reading which student's shutdown is grief versus defiance, and running a 45-minute lab with 28 bodies and one set of scissors has no digital substitute, holding it at 13.
Hands-on in uncontrolled environments Cafeteria and hallway duty, bus dismissal, breaking up a shoving match, walking a nosebleed to the nurse, and chaperoning field trips put you in unscripted physical space with minors daily — short of trades work in an active hazard zone, which is why this sits at 15 and not 18.
Licensed human required and personally liable State teaching licence with subject endorsement and periodic renewal, plus personal mandatory-reporter exposure under state child-abuse statutes and in-loco-parentis duty of care, means the credential is legally required for the room — 12 rather than higher because district and state immunity absorbs most negligence claims and you are rarely the named defendant.
Meaningful discretion You decide in real time whether a behavior is a warning, a call home, or an office referral that starts a discipline record, and whether a comment about home warrants a report — real discretion, but bounded by the district code of conduct, IEP/504 accommodations you must follow, and pacing guides, which is why this lands at 13 rather than in the ownership band.
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 (13/20 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 (12/20) is whether the law requires a licensed human to sign. Trust premium (18/20) 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 43 of this occupation's 71 points (61%).
Embodiment (15/20) 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 84/100, still SAFE.
State legislatures or boards of education codifying that AI may not be the instructor of record and that a licensed teacher must supervise and sign off on any AI-generated instruction, IEP-adjacent accommodation, or grade of record — the pattern in bills already filed in states like Tennessee, Ohio and California on AI use in K-12, plus existing FERPA/COPPA data-handling duties assigned to a named certificated employee
Task-mix shift: as AI absorbs lesson drafting, quiz banks and rubric grading, the residual job concentrates in the judgment tier — behavior escalation, retention and placement recommendations, MTSS/504 referrals, custody and safety calls — raising the share of the day spent on consequential ambiguous decisions
Ratio and class-size floors written into state law or union contract (e.g. CTU/UTLA-style bargaining over class size and 'no AI substitute for staffing'), which converts parental preference for human teachers into an enforced headcount rather than a sentiment
Expansion of mandatory-reporter and student-safety liability: statutes or district policy making the supervising teacher explicitly responsible for detecting AI-flagged self-harm/threat signals and acting, so the human is the accountable link in the chain (already emerging in threat-assessment team requirements after school-safety laws in TX, FL)
Same task-mix shift raises the resistant share: if prep and grading are automated, what remains (de-escalation, adolescent motivation, live differentiation for 25 bodies) is the part AI cannot do — though this can also cut hours per teacher rather than raise the score
The limit. Embodiment is already near its practical maximum for a supervised-minors role and cannot rise. The main downside risk is not AI capability but staffing policy: larger classes plus AI tooling reduce headcount without changing any dimension score.
| New York-Newark-Jersey City, NY-NJ | 39,310 | $96,500 +50% |
| Dallas-Fort Worth-Arlington, TX | 21,590 | $65,660 +2% |
| Houston-Pasadena-The Woodlands, TX | 19,340 | $65,830 +2% |
| Chicago-Naperville-Elgin, IL-IN | 17,300 | $78,000 +21% |
| Washington-Arlington-Alexandria, DC-VA-MD-WV | 14,890 | $77,350 +20% |
| Atlanta-Sandy Springs-Roswell, GA | 13,650 | $75,490 +17% |
| Philadelphia-Camden-Wilmington, PA-NJ-DE-MD | 13,090 | $78,130 +21% |
| Los Angeles-Long Beach-Anaheim, CA | 12,780 | $98,370 +53% |
| Mount Vernon-Anacortes, WA | 160 | $106,920 +66% |
| San Jose-Sunnyvale-Santa Clara, CA | 3,330 | $105,030 +63% |
| Olympia-Lacey-Tumwater, WA | 420 | $104,410 +62% |
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 71. 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.