EXPOSED
The documentation core of this job — writing requirements specs, drawing process flows, building cost-benefit spreadsheets, translating business asks into technical stories, drafting test plans and user guides — is exactly what LLMs already do at usable quality. What survives is the political and organizational work: sitting in a room with a finance director and a warehouse supervisor who disagree about what 'inventory' means, and owning the call on scope, sequencing, and vendor selection. No license, no signature requirement, so the only moats are stakeholder trust and accountability for whether the system actually gets adopted.
This fall is concentrated in 2020 and has not recovered since.
Median pay $90,920 → $105,850 -6.9% 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
+8.7% 521,100 → 566,500 on the projections basis
Exposed, but growing
AI can already do a lot of these tasks, and the BLS still expects +8.7% 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.
~34,200 openings a year on average, including replacing people who leave.
InformaticistCoding AuditorNurse ReviewerRecords ManagerSystems AnalystComputer AnalystDatabase AnalystInternet AnalystSoftware AnalystTelehealth NurseInformatics NurseTechnical AnalystComputer ArchitectFunctional AnalystHealthcare AnalystPhysician ReviewerProgrammer AnalystSolution ArchitectSystems IntegratorTechnology AnalystApplication AnalystClinical SpecialistComputer ConsultantNurse Informaticist
Holding it up: judgment & accountability . Weakest point: liability shield .
Mixed — a routine tier and a judgment tier Requirements documents, as-is/to-be process diagrams, UAT scripts, data-mapping tables and gap analyses are structured text generation that models produce at first-draft quality, which caps this at 9; what pulls it above the 6 line is elicitation itself — discovering that the shipping team maintains a shadow Excel workbook nobody mentioned in the intake meeting, and running the workshop where three departments are forced to agree on a single definition of 'active customer'.
Fully desk- and screen-based A 4 reflects a job done entirely in Jira, Visio, Confluence and Teams calls, with the only physical element being occasional presence on a plant floor or in a clinic to watch how the current system is actually used — observation you could do over video and often do.
No licence, no signature requirement No state licenses systems analysts, nothing you produce carries a stamp, and when a requirement is missed the vendor contract or the CIO absorbs it; the 2 rather than 0 acknowledges that CBAP/PMP or vendor certifications appear on job postings and occasionally on SOW staffing requirements, but they gate nobody's work.
Meaningful discretion The 12 sits at the top of real discretion: you decide what goes in release one versus the backlog, whether to configure or customize, and which vendor demo actually meets the requirement — calls that cost millions and get overruled by a steering committee, which is exactly why it isn't 15.
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 (2/20) is whether the law requires a licensed human to sign. Trust premium (10/20) is whether buyers specifically pay for a person. Judgment and accountability (12/20) 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 24 of this occupation's 37 points (65%).
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 computer systems analysts 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 51/100, still EXPOSED.
If a named individual must be designated as accountable for AI system deployment decisions — as under EU AI Act Art. 26 deployer obligations, Colorado SB 24-205, and NYC Local Law 144 bias-audit duties — analysts specifying and selecting AI-containing systems become the documented owner of suitability and impact-assessment calls.
As spec-drafting, story-writing and test-plan generation get fully absorbed by tools like Copilot/Devin-style agents, the residual job becomes elicitation under conflicting stakeholder definitions, legacy data-model archaeology, and cutover sequencing — work that requires undocumented institutional knowledge no model has ingested. This is a genuine two-tier job and the judgment tier is what remains.
If cyber-insurance underwriters or big-4 audit engagement letters begin requiring that requirements and vendor-selection work for material systems be performed by a named human consultant rather than AI-generated artifacts — mirroring how insurers now demand named MFA/EDR owners — clients pay specifically for attributable human authorship.
If federal procurement or FedRAMP/CMMC-style rules require a named systems architect of record to attest to requirements traceability and security control mapping (analogous to the P.E. stamp sought in some software-engineering licensure bills, e.g. Texas's now-retired software PE), a signature obligation appears where none exists.
The limit. Even with all of these, the occupation stays mid-band: no existing license, low physicality, and the documentation core is already commoditized. Realistic ceiling is roughly the high 40s/low 50s, and only for analysts working on regulated or safety-adjacent systems — internal IT analysts at ordinary firms have no route to the liability or insurer levers.
| New York-Newark-Jersey City, NY-NJ | 26,540 | $128,340 +21% |
| Dallas-Fort Worth-Arlington, TX | 20,000 | $122,520 +16% |
| Seattle-Tacoma-Bellevue, WA | 19,050 | $133,690 +26% |
| Washington-Arlington-Alexandria, DC-VA-MD-WV | 18,620 | $127,040 +20% |
| Boston-Cambridge-Newton, MA-NH | 17,470 | $129,180 +22% |
| Los Angeles-Long Beach-Anaheim, CA | 16,900 | $121,560 +15% |
| Chicago-Naperville-Elgin, IL-IN | 14,660 | $103,490 -2% |
| San Francisco-Oakland-Fremont, CA | 11,180 | $134,460 +27% |
| San Jose-Sunnyvale-Santa Clara, CA | 7,260 | $157,140 +48% |
| Pueblo, CO | 40 | $139,870 +32% |
| San Francisco-Oakland-Fremont, CA | 11,180 | $134,460 +27% |
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 37. 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.