EXPOSED
The bulk of this job — reading policy language against a loss description, coding estimates from photos, computing depreciation, drafting denial and settlement letters, flagging fraud indicators — is exactly the document-and-pattern work insurers are already routing through automated straight-through processing and photo-based estimating tools. What holds is the field tier: climbing roofs after a hailstorm, scoping fire and water damage in person, taking recorded statements, and the named human who owns a large or contested settlement. Adjuster licensing exists in most states but is a low bar and does not require personal sign-off the way an engineering stamp or medical license does.
Dipped in 2020, then grew past where it started.
Median pay $66,790 → $78,000 -6.6% 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
-5.1% 356,100 → 337,900 on the projections basis
Exposed, and shrinking
Both signals point the same way: the tasks are largely automatable and the BLS projects -5.1% 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.
~21,100 openings a year on average, including replacing people who leave.
AdjusterClaim AgentInvestigatorFire AdjusterClaims AnalystClaims AuditorClaim InspectorClaims AdjusterClaims AdjustorClaims ApproverClaims ExaminerAdjustment ClerkClaims ProcessorGeneral AdjusterClaims ConsultantClaims SpecialistClerical AdjusterInsurance AuditorProperty AdjusterBenefit AuthorizerDisability AnalystField InvestigatorInsurance AdjusterInsurance Examiner
Holding it up: judgment & accountability . Weakest point: task resistance .
Core tasks are already automatable Xactimate line-item estimating from claimant-uploaded photos, auto total-loss valuations pulled from ACV databases, ISO ClaimSearch fraud hits, and templated reservation-of-rights and denial letters are already running straight-through at carriers like Lemonade and Progressive with no adjuster touch, which is why this sits at 6 rather than mid-band — the residual field inspection work is a minority of claim volume, not the core.
Some physical or field component A 7 reflects the split desk: property and catastrophe adjusters do ladder-and-roof scoping, crawl-space water inspections, and salvage yard vehicle teardowns in storm conditions, but auto desk examiners, workers' comp examiners, and life/health claims staff work entirely from a queue and never leave the building.
Certification preferred, not legally required Most states license adjusters through a 40-hour course and multiple-choice exam with reciprocity across state lines, and Texas and Florida let unlicensed trainees work under a sponsor — the carrier, not you, is the party sued for bad faith under unfair claims settlement practices acts, so your name on the file carries almost no personal exposure, unlike a public adjuster's fiduciary duty to the insured.
Meaningful discretion Coverage determinations, causation calls on wind-versus-flood, and reserve setting are real discretion, but they run inside authority limits — anything above your dollar threshold escalates to a supervisor or claims committee, coverage questions go to counsel opinion, and your file must document the reasoning against company guidelines, which caps this at 9.
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 (6/20 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 (7/20) is whether the law requires a licensed human to sign. Trust premium (7/20) is whether buyers specifically pay for a person. Judgment and accountability (9/20) 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 23 of this occupation's 36 points (64%).
Embodiment (7/20) 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 56/100, still EXPOSED.
Task-mix shift: if straight-through processing absorbs the low-severity, undisputed tier (already visible in Lemonade/Tractable-style auto and property flows), the residual role is complex-loss, coverage-dispute, reservation-of-rights, subrogation and suspected-fraud work where the adjuster owns a contested call under ambiguity and is deposed on it. Fewer jobs, but each one scores higher on this dimension.
Same two-tier shift raises task resistance for the survivors: recorded statements under conflicting witness accounts, causation disputes needing an engineer/contractor argued against, reconstructing intent in fraud investigations. Note this is a composition effect within the surviving field/complex tier, not a capability regression.
Litigation outcomes treating fully automated denial as per se bad faith — the Cigna PXDX and UnitedHealth nH Predict class actions, and California SB 1120 (2024) barring AI-only medical necessity denials without licensed-clinician review. If SB 1120's logic is extended by statute beyond health utilization review to property/casualty adverse determinations, a licensed sign-off becomes structural rather than optional.
State insurance departments adopting rules that require a named licensed adjuster to personally review and sign any adverse claim determination produced with algorithmic assistance — the direction of the NAIC Big Data and AI Working Group model bulletin (adopted in ~20 states) plus Colorado's SB21-169 quantitative-testing regime. A hard 'licensed human decision-maker on every denial' requirement, as opposed to today's governance-only language, plus market-conduct exam exposure and bad-faith liability attaching to the signer, would move this materially.
Catastrophe-response requirements — state DOI emergency adjuster licensing after hurricanes/hail, plus reinsurer and carrier demands for physical scoping on large-loss and total-loss claims where drone/photo estimates are contested in appraisal or litigation. If carriers formalize a severity threshold above which an in-person scope is mandatory, on-site attendance becomes a documented step rather than a discretionary one.
The limit. No realistic route to a higher trust premium: the buyer of adjusting is the carrier, not the claimant, and the carrier's incentive is cost per claim. Claimants who want a human hire public adjusters or attorneys — a different occupation. Also note the liability and judgment levers concentrate value in a much smaller complex-loss and field cohort; the desk examiner tier is not protected by any of the above.
| New York-Newark-Jersey City, NY-NJ | 14,840 | $95,980 +23% |
| Atlanta-Sandy Springs-Roswell, GA | 13,830 | $75,770 -3% |
| Phoenix-Mesa-Chandler, AZ | 11,350 | $64,370 -17% |
| Chicago-Naperville-Elgin, IL-IN | 9,080 | $78,640 +1% |
| Dallas-Fort Worth-Arlington, TX | 8,630 | $82,890 +6% |
| Los Angeles-Long Beach-Anaheim, CA | 7,540 | $89,070 +14% |
| Tampa-St. Petersburg-Clearwater, FL | 7,240 | $77,990 +0% |
| Philadelphia-Camden-Wilmington, PA-NJ-DE-MD | 6,850 | $81,560 +5% |
| Merced, CA | 40 | $110,910 +42% |
| Burlington-South Burlington, VT | 70 | $108,510 +39% |
| San Francisco-Oakland-Fremont, CA | 3,390 | $105,340 +35% |
Hollard · UnitedHealth · Chubb
Insurance Business reports that insurer Hollard is testing AI to speed up consultants' reviews of claims files.
Insurer Chubb reportedly announced a 20% staff reduction alongside adoption of AI systems for claims processing and underwriting.
Healthcare Finance News reports a class action lawsuit over UnitedHealth's use of an AI algorithm (nH Predict) to deny post-acute care claims has been allowed to proceed, confirming the insurer's use of automated claim review in place of human adjudication.
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.