EXPOSED
The modal worker here presides over benefits, licensing, workers' comp, and immigration-type hearings, weighs testimony and documentary evidence, and issues written findings of fact and conclusions of law — the decision-writing half of that is exactly what LLMs draft well from a record, and agencies under backlog pressure are already piloting AI-assisted decision drafting and evidence summarization. What holds is the constitutional and statutory requirement that a neutral human adjudicator hear the case, assess witness credibility in real time, and personally sign the order that strips or grants someone's benefits or license. Note the title bundles bar-licensed federal ALJs (strong shield) with state hearing officers and claims adjudicators who need no license (much thinner shield); the routine-docket adjudicator tier is where headcount compresses first.
Dipped in 2020, then grew past where it started.
Median pay $97,870 → $117,860 -3.7% 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
-0.7% 17,500 → 17,400 on the projections basis
Shrinking, but not obviously because of AI
The BLS projects -0.7% by 2034, but at 58/100 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.
~500 openings a year on average, including replacing people who leave.
AdjudicatorAppeals OfficerAppeals RefereeCounty OrdinaryHearing OfficerAppeals ExaminerHearing ExaminerAppellate ConfereeClaims AdjudicatorHousing Court JudgeAdministrative JudgeJustice of the PeaceField Hearing OfficerTraffic Court RefereeParole Hearing OfficerAdjudications SpecialistAdministrative Law JudgeVeteran Appeals ReviewerClinical Appeals ReviewerDisability Hearing OfficerLegal Activity AdjudicatorDisciplinary Hearing OfficerChild Support Hearing OfficerAdministrative Hearing Officer
Holding it up: judgment & accountability . Weakest point: embodiment .
Mixed — a routine tier and a judgment tier Roughly half the day is record review and drafting findings of fact and conclusions of law from a documentary file — SSA disability decisions built off medical exhibits and vocational grids are template-driven enough that draft-generation tools already touch them — while live credibility assessment, ruling on evidentiary objections mid-hearing, and questioning an unrepresented claimant who contradicts himself are not, which is what keeps this at 10 and not 5.
Fully desk- and screen-based The work is a hearing room, a video teleconference link, a case file, and an order — physical presence is required by statute in some venues but nothing about the task depends on the body doing it, hence 4 rather than 0 only because in-person and site-visit hearings still exist in workers' comp and land-use matters.
Licensed human required and personally liable Federal ALJs under 5 U.S.C. 556-557 are bar-licensed, appointed under the Appointments Clause after Lucia, and personally sign orders subject to appeal and judicial review — that is a real named-human accountability structure; it sits at 15 rather than 19 because a large share of state hearing officers and unemployment/benefits claims adjudicators hold the title with no bar admission and no licence to revoke.
Exists to be accountable for ambiguous calls You decide whether a witness is lying, whether an impairment meets a listing, whether a licence to practice gets pulled — on incomplete records, often with an unrepresented party on one side, and the order takes effect on your signature with only appellate review behind it; 17 not 20 because regulations, benefit schedules, and precedential agency decisions constrain the outcome space more than a trial judge faces.
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 (10/20 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 (15/20) is whether the law requires a licensed human to sign. Trust premium (12/20) is whether buyers specifically pay for a person. Judgment and accountability (17/20) 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 44 of this occupation's 58 points (76%).
Embodiment (4/20) 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.
No occupation passed every test: close enough to administrative law judges, adjudicators, and hearing officers on skills and subject matter, at least 10 points more resistant, no big jump in training, no new licence, no pay cut, and not shrinking on its own. That happens for 223 of the 654 occupations here that aren't SAFE, and it is worth stating plainly rather than leaving the section off.
The usual reason is that exposure travels with the skill profile. The jobs most similar to yours tend to be exposed for the same reasons yours is, so the near neighbours don't clear the gap — and the ones that do are a different kind of work, not a transfer of what you already know. Read that as a limit of this method, not a verdict that you're stuck: it only compares whole occupations, and it cannot see specialisation, industry, or anything you'd bring that isn't in a federal skill survey.
Here is that claim on your own job rather than in the abstract. These are the three occupations closest to this one by skill and subject matter — the places the work would most naturally transfer — with what the register scores them:
That is the whole problem in three lines. The nearest work is not meaningfully safer, so there is no move here that trades a similar skill set for a better verdict. This is not us running out of ideas — it is what the neighbourhood looks like.
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 68/100 — SAFE.
Genuine two-tier structure: if evidence summarization and boilerplate conclusions-of-law drafting are fully absorbed, the residual day is live credibility assessment, pro se claimant colloquy, and contested-record cases where the parties dispute what the record even says. Headcount falls but the surviving role's task mix is harder to automate.
An explicit APA amendment or state UAPA analogue barring AI-generated findings of fact and credibility determinations without a named ALJ's personal certification that they independently reviewed the record — the model already exists in state court rules (e.g., Texas 5th Circuit-style AI certification orders, Illinois Supreme Court AI policy 2025) and in the 2024 Social Security Administration OIG scrutiny of AI drafting tools. A statutory personal-certification duty with sanction exposure would harden the thin state-hearing-officer tier toward the federal ALJ standard.
Licensure creep: state administrative procedure acts requiring bar admission for hearing officers who issue final agency orders (already true in some states for workers' comp judges). This converts the unlicensed claims-adjudicator tier into a signature-bearing profession.
Due-process litigation outcomes — a Mathews v. Eldridge-line ruling that AI-assisted drafting in benefits terminations violates the right to a decision by the officer who heard the evidence — would relocate accountability firmly onto the individual adjudicator rather than the agency's pipeline.
The limit. Trust premium has no realistic route up: parties do not choose their adjudicator and cannot pay for a human one, so demand-side preference cannot register. Embodiment is capped by remote-hearing normalization post-2020. The task_resistance rise here is a compositional artifact of headcount loss, not protection for the current workforce.
| New York-Newark-Jersey City, NY-NJ | 1,230 | $130,960 +11% |
| Washington-Arlington-Alexandria, DC-VA-MD-WV | 660 | $151,990 +29% |
| Houston-Pasadena-The Woodlands, TX | 420 | $124,400 +6% |
| Atlanta-Sandy Springs-Roswell, GA | 390 | $72,440 -39% |
| Columbus, OH | 390 | $77,330 -34% |
| Philadelphia-Camden-Wilmington, PA-NJ-DE-MD | 390 | $95,350 -19% |
| Chicago-Naperville-Elgin, IL-IN | 350 | $121,720 +3% |
| Little Rock-North Little Rock-Conway, AR | 320 | $64,620 -45% |
| Tallahassee, FL | 60 | $177,150 +50% |
| Kansas City, MO-KS | 60 | $164,530 +40% |
| Washington-Arlington-Alexandria, DC-VA-MD-WV | 660 | $151,990 +29% |
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 58. 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.