EXPOSED
The paper-facing half of this job — syllabus construction, lecture notes, reading guides, rubric-based comments on freshman comp essays, plagiarism triage — is exactly what language models do cheaply, and the writing assignments themselves are being destabilized by student AI use. The classroom half survives: running a live seminar where twenty people argue about a text, coaching a student through four drafts, writing the recommendation letter that carries a name behind it. The bigger near-term threat isn't AI replacing professors, it's collapsing English enrollment and adjunct-heavy staffing shrinking the seat count, and there is no license protecting the role.
This fall is concentrated in 2020 and has not recovered since.
Median pay $68,490 → $78,760 -8.0% 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
0%
Percentage only. The projection counts a different population from the 57,720 above — it includes self-employed workers, which for this occupation is most of them, so the two headcounts are not comparable.
Shrinking, but not obviously because of AI
The BLS projects 0% by 2034, but at 46/100 this work is only moderately exposed — not the profile of a job current AI can simply do. Occupations shrink for many reasons, and the score does not point at automation as this one's cause.
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.
~5,100 openings a year on average, including replacing people who leave.
TeacherLecturerProfessorInstructorFaculty MemberEnglish AdjunctEnglish TeacherAdjunct LecturerClassics TeacherEnglish LecturerAdjunct ProfessorCollege ProfessorEnglish ProfessorEtymology TeacherReading ProfessorAdjunct InstructorEnglish InstructorReading InstructorAssistant ProfessorAssociate ProfessorComposition TeacherEtymology ProfessorLanguage InstructorHumanities Professor
Holding it up: trust premium . Weakest point: liability shield .
Mixed — a routine tier and a judgment tier Grading a stack of comp essays against a rubric, building reading lists, and drafting lecture notes on Beloved are all things a model does in seconds, but a two-hour seminar where you read the room, redirect a student who has misread the poem's tone, and push a discussion past the obvious reading is not scriptable — roughly half the workload holds, which is why this sits at 11 rather than in the resistant teens.
Some physical or field component You are physically in a room three days a week — projecting to the back row, watching who has stopped taking notes, holding office hours where a student cries about their thesis — but the room is a climate-controlled classroom and everything you produce could be emailed, so this lands at 8 rather than the 14+ of work done outdoors or on machinery.
No licence, no signature requirement No state licenses postsecondary English instruction: a PhD or MFA is a hiring credential, not a legal gate, and adjuncts are routinely hired on a master's with a semester's notice — nobody's signature on a syllabus creates personal liability, hence 2.
Meaningful discretion You decide whether a paper is plagiarized or AI-generated and what that does to a student's transcript, whether a student who missed six weeks passes, and what a contested text is worth on the syllabus — genuine discretion with consequences, but appealable through department chairs and academic integrity committees rather than resting on you alone, which is what separates 11 from 16.
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 (14/20) is whether buyers specifically pay for a person. Judgment and accountability (11/20) 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 27 of this occupation's 46 points (59%).
Embodiment (8/20) 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.
Law Teachers, Postsecondary EXPOSED
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 60/100, still EXPOSED.
If departments complete the shift already underway from take-home essays to assessment forms AI can't ghostwrite — in-class handwritten timed writing, oral defenses of drafts, one-on-one draft conferences, portfolio process documentation — the automatable tier (rubric comments on submitted prose) leaves the job and what remains is live, embodied coaching. This is a genuine two-tier occupation and the routine tier is the one being eaten.
If AI-use allegations continue to route through faculty as the originating accuser in academic integrity cases — where the instructor's judgment call, not a detector score, is what an honor board can act on, and where several universities have already barred Turnitin AI scores as sole evidence — the role formally owns a consequential, appealable, sometimes litigated determination about a student's record.
If the US Department of Education's 'regular and substantive interaction' requirement for distance education (34 CFR 600.2) is enforced or tightened to explicitly exclude AI-generated feedback and require an identified instructor of record, the human presence becomes a condition of Title IV aid eligibility rather than a preference — that is a purchasing rule, not sentiment.
If accreditor faculty-credential rules (e.g. SACSCOC's 18 graduate semester hours in the discipline) are read to require a named credentialed human as instructor of record who signs the grade roster and cannot delegate that to a system, the role acquires a weak quasi-license anchor. Note this is institutional, not personal legal liability, so the ceiling here is low.
If faculty union contracts add clauses reserving grading and course-content authorship to bargaining-unit members — language of the type CFA, PSC-CUNY, and several 2024-25 grad-worker contracts have sought on AI and workload — the human is contractually required for a defined slice of the work.
The limit. Every lever here protects the role's content, not its headcount. The binding constraint is English enrollment decline and adjunctification: a course can be fully human-taught, human-graded, accreditor-compliant and still not exist. Raising these dimensions changes what the surviving professor does, not how many survive.
| New York-Newark-Jersey City, NY-NJ | 5,320 | $96,430 +22% |
| Los Angeles-Long Beach-Anaheim, CA | 2,150 | $128,810 +64% |
| Boston-Cambridge-Newton, MA-NH | 2,110 | $87,020 +10% |
| Chicago-Naperville-Elgin, IL-IN | 1,850 | $78,620 +0% |
| Philadelphia-Camden-Wilmington, PA-NJ-DE-MD | 1,220 | $83,500 +6% |
| Washington-Arlington-Alexandria, DC-VA-MD-WV | 1,030 | $82,080 +4% |
| Houston-Pasadena-The Woodlands, TX | 940 | $102,500 +30% |
| Dallas-Fort Worth-Arlington, TX | 920 | $81,570 +4% |
| San Francisco-Oakland-Fremont, CA | 610 | $165,090 +110% |
| Riverside-San Bernardino-Ontario, CA | 480 | $160,050 +103% |
| Fresno, CA | 200 | $137,730 +75% |
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.