EXPOSED
The analytical core of this job — variance analysis, cash-flow forecasts, budget consolidation, monthly reporting packets, covenant tracking — is exactly the structured text-and-spreadsheet work AI now does at usable quality, and the reporting layer under a financial manager is shrinking fast. What survives is accountability: someone must own the forecast when it's wrong, negotiate with lenders and auditors, approve capital allocation under genuine ambiguity, and answer to a board. The title bundles controllers, treasury managers, and bank branch/lending managers; the modal worker is a corporate controller-type whose staff headcount falls faster than their own seat does.
Headcount grew steadily across the period.
Median pay $129,890 → $166,570 +2.6% 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
+14.8% 868,600 → 997,400 on the projections basis
Growing, and only partly exposed
The BLS expects +14.8% more of these jobs by 2034, and at 48/100 the work is only partly exposed — some tasks are automatable, the core of the job is not. Nothing here is in tension.
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.
~74,600 openings a year on average, including replacing people who leave.
FinancierPaymasterTreasurerControllerComptrollerTax ManagerBank ManagerCash ManagerTax DirectorAsset ManagerTrust OfficerBranch ManagerBudget ManagerCity TreasurerCredit ManagerFiscal ManagerWealth ManagerAccount ManagerBanking ManagerCity ControllerCost ControllerFinance ManagerAuditing ManagerCity Comptroller
Holding it up: judgment & accountability . Weakest point: embodiment .
Mixed — a routine tier and a judgment tier Building the consolidated month-end package, rolling the 13-week cash forecast, tying out intercompany eliminations and recalculating covenant ratios are all deterministic transformations of ledger data that current tools reproduce at speed, which pins this at 9 rather than higher — what pulls it up off the floor is the un-automatable half: sitting across from a lender renegotiating a leverage covenant, defending a reforecast to the audit committee, and deciding which department's budget request dies this quarter.
Fully desk- and screen-based The work happens in the ERP, the model, and the conference room; the only physical elements are inventory or cash counts you observe rather than perform and site visits to a plant or branch, which is why this sits at 3 and not 0 — nothing about the job requires your hands.
Certification preferred, not legally required A CPA or CFA is common and often preferred but not statutorily required to hold the controller seat, and the personal exposure that does exist is real but narrow — SOX 302/906 sub-certifications, and for the treasury and bank-lending variants under this SOC, signing authority on filings and credit decisions — so 8 reflects genuine but conditional personal liability rather than the licence-or-you-don't-practise structure of a public auditor or actuary.
Exists to be accountable for ambiguous calls At 16 because capital allocation, revenue-recognition judgment calls, reserve and impairment estimates, and going-concern assessments are made on incomplete information with no procedure that dictates the answer, and the consequence — a restatement, a covenant breach, a misallocated capex cycle — lands on your signature and your board minutes, not on the analyst who built the model.
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 (8/20) 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 (16/20) 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 36 of this occupation's 48 points (75%).
Embodiment (3/20) 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 financial managers 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 62/100, still EXPOSED.
Genuine two-tier structure: if the close, consolidation, variance commentary and covenant tracking tier is automated, the residual seat is materially all lender/auditor negotiation, capital allocation under ambiguity, reforecasting in a shock, and defending numbers to a board — work with no verifiable ground truth at decision time. Score rises by attrition of the routine tier, not by new capability limits. Watch for controller job postings that no longer list close-cycle duties.
Lender and insurer paper rather than statute: credit agreements and D&O/fidelity policies requiring covenant compliance certificates and borrowing-base certificates be signed by a named, individually identified officer, with carriers excluding losses from unreviewed automated reporting. Already visible in private-credit covenant certificate language and in cyber/crime policy exclusions for automated payment authorization.
SEC extending sub-certification of internal controls beyond the CEO/CFO — i.e. a rule or an auditor-driven practice requiring the named controller to personally attest that AI-generated figures in the 10-Q/10-K were reviewed by a human. Adjacent to this: PCAOB's technology-assisted analysis standard (effective 2025) pushing auditors to demand a named human owner of each automated close control, and SEC enforcement of 'AI-washing' in disclosures. A formal sub-certification requirement is the specific thing to watch.
Little headroom — already 16. Would move only if boards formally delegate named AI-model-risk ownership to the finance chair, e.g. an SR 11-7-style model risk management expectation extended from banks to non-bank corporate finance functions, naming an accountable officer for finance models.
The limit. Trust premium has no realistic route: buyers of a controller's output are internal (board, lenders, auditors) and pay for signature and accountability, not for human authorship — so any gain shows up in liability_shield, not here. Embodiment is structurally fixed near zero. Realistic ceiling is roughly the high 50s, and it is almost entirely a legal-signature story: absent a sub-certification or model-risk rule that names the controller, the score drifts down as the analytical tier automates and the headcount beneath the seat disappears.
| New York-Newark-Jersey City, NY-NJ | 89,960 | $221,010 +33% |
| Chicago-Naperville-Elgin, IL-IN | 41,590 | $168,810 +1% |
| Los Angeles-Long Beach-Anaheim, CA | 34,480 | $175,310 +5% |
| Dallas-Fort Worth-Arlington, TX | 26,600 | $169,200 +2% |
| Washington-Arlington-Alexandria, DC-VA-MD-WV | 26,430 | $185,270 +11% |
| Boston-Cambridge-Newton, MA-NH | 21,650 | $209,780 +26% |
| Houston-Pasadena-The Woodlands, TX | 18,570 | $166,910 +0% |
| Philadelphia-Camden-Wilmington, PA-NJ-DE-MD | 18,520 | $166,920 +0% |
| San Jose-Sunnyvale-Santa Clara, CA | 7,970 | $225,280 +35% |
| New York-Newark-Jersey City, NY-NJ | 89,960 | $221,010 +33% |
| San Francisco-Oakland-Fremont, CA | 17,040 | $216,150 +30% |
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 48. 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.