SAFE
The core work is physical and unforgiving: intubating airways, placing central lines and epidurals, titrating agents minute-by-minute against a bleeding, hypotensive, unpredictable patient. Closed-loop sedation algorithms and monitoring alarms already exist and will keep improving, but a licensed physician must be present and personally liable for induction, crisis rescue, and extubation. The automatable slice is documentation, pre-op risk stratification from chart data, and dosing suggestions — real efficiency gains, not job replacement.
Dipped in 2020, then grew past where it started.
Median pay $239,200 → $391,490 +30.9% 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
+3.2%
Percentage only. The projection counts a different population from the 38,760 above — it includes self-employed workers, which for this occupation is most of them, so the two headcounts are not comparable.
Hard to automate, and growing
The work resists current AI and the BLS projects +3.2% more of these jobs by 2034. Note that safe does not mean well paid — several of the fastest-growing resistant occupations are among the lowest paid on the register.
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.
~1,300 openings a year on average, including replacing people who leave.
AnesthetistAnesthesiologistAnaesthesiologistStaff AnesthetistMedical Doctor (MD)Staff AnesthesiologistAnesthesiology PhysicianCardiac AnesthesiologistGeneral AnesthesiologistPediatric AnesthesiologistAmbulatory AnesthesiologistObstetrical AnesthesiologistAnesthesiology Pain Management PhysicianDO Physician (Doctor of Osteopathic Medicine Physician)
Holding it up: liability shield . Weakest point: trust premium .
Tasks largely resist digitisation At 17 rather than 20, the genuinely digitisable slice is real — ASA physical status scoring from chart data, pre-op checklists, and automated anesthesia record charting are already shipping — but airway rescue on a Mallampati IV patient, awake fiberoptic intubation, and hand-titrating pressors through a hemorrhage are not tasks a model can do, and closed-loop propofol systems like McSleepy still require a physician at the head of the bed.
Hands-on in uncontrolled environments A 19 reflects that nearly every minute is physical and in an uncontrolled field: laryngoscopy in a patient whose neck won't extend, ultrasound-guided internal jugular access, loss-of-resistance for an epidural in a laboring patient who is moving, plus trauma bays, OB suites, and out-of-OR sedation in radiology where the equipment and lighting are whatever you find.
Licensed human required and personally liable 20 is the ceiling because you hold an unrestricted state medical licence and ABA board certification, sign the anesthesia record yourself, and wrongful-death and anoxic-brain-injury claims name you personally — this is the specialty that rebuilt its own malpractice profile through the ASA Closed Claims Project precisely because the physician, not the device, is the defendant.
Exists to be accountable for ambiguous calls 19 because the calls are yours alone and made in seconds: cancel the case for an unexplained troponin, convert to general when the block fails at incision, declare a can't-intubate-can't-ventilate emergency and cut the neck, decide whether a hypotensive patient on a MAOI gets ephedrine or phenylephrine — no protocol resolves these and there is no time to consult upward.
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 (17/20 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 (20/20) 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 (19/20) 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 53 of this occupation's 89 points (60%).
Embodiment (19/20) 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 94/100, still SAFE.
Task-mix shift as closed-loop TIVA/TCI systems and automated documentation (Epic Anesthesia, autocharting) absorb the routine ASA I-II maintenance tier, leaving the day dominated by high-acuity cases: cardiac, obstetric hemorrhage, difficult airway, pediatric, and pre-op optimization of ambiguous multi-morbid patients. Also chronic pain and critical care sidelines expand the judgment tier.
Continued growth of patient-selected 'physician-only anesthesia care' — explicit MD/DO presence guarantees marketed by hospitals and demanded in surgical consent conversations, plus ASA's ongoing campaign against opt-out of physician supervision for CRNAs. If state opt-out rollbacks (as ASA has pushed in several state legislatures) succeed, buyer-side preference for a physician anesthesiologist becomes contractually locked rather than merely preferred.
The limit. At 89/100 with liability_shield and embodiment effectively saturated, only trust_premium has real headroom, and that headroom is contested by the CRNA scope-of-practice fight rather than by AI. The realistic risk direction for this occupation is scope substitution by mid-level clinicians augmented by decision support, not model capability.
| New York-Newark-Jersey City, NY-NJ | 7,080 | $460,800 +18% |
| Chicago-Naperville-Elgin, IL-IN | 1,700 | $77,770 -80% |
| Houston-Pasadena-The Woodlands, TX | 1,160 | $210,230 -46% |
| Seattle-Tacoma-Bellevue, WA | 990 | $646,930 +65% |
| Dallas-Fort Worth-Arlington, TX | 950 | $218,060 -44% |
| Boston-Cambridge-Newton, MA-NH | 860 | — |
| Atlanta-Sandy Springs-Roswell, GA | 770 | $226,830 -42% |
| Orlando-Kissimmee-Sanford, FL | 700 | $239,200 -39% |
| Seattle-Tacoma-Bellevue, WA | 990 | $646,930 +65% |
| St. Cloud, MN | 40 | $591,740 +51% |
| Louisville/Jefferson County, KY-IN | 500 | $491,640 +26% |
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 89. 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.