SAFE
The core of the job is physically leading adaptive sports, arts, community outings, and social-skills groups with patients who have brain injuries, psychiatric conditions, or mobility limits — that is in-room, hands-on work AI cannot do. What AI does absorb is the paperwork layer: treatment plan drafting, progress notes, insurance documentation, and activity ideation. Credentialing (CTRS) plus state licensure in a handful of states gives partial protection, but the stronger moats are embodiment and the therapeutic relationship itself.
Part 2020 shock, part continued decline in the years since.
Median pay $48,220 → $61,960 +2.8% 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
+3.3% 16,100 → 16,600 on the projections basis
Hard to automate, and growing
The work resists current AI and the BLS projects +3.3% 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.
TherapistEcotherapistDance TherapistDrama TherapistActivity TherapistLifestyle DirectorTherapeutic MentorAdventure TherapistActivities TherapistRecreation TherapistActivities CoordinatorCreative Art TherapistRecreational TherapistTherapeutic SpecialistHorticultural TherapistRehabilitation TherapistGeneral Activities TherapistTherapeutic Riding InstructorMusic Rehabilitation TherapistRecreational Therapy TechnicianCertified Recreational TherapistLife Enrichment Coordinator (LEC)Therapeutic Recreation SpecialistRecreational Therapy Program Coordinator
Holding it up: trust premium . Weakest point: liability shield .
Tasks largely resist digitisation Running a wheelchair basketball session, transferring a stroke patient into an adaptive kayak, or de-escalating a psychiatric inpatient mid-group are tasks that stay human, but the Functional Independence Measure scoring, CMS-required treatment plan templates, activity-idea generation, and discharge summaries are a real slice of the week that software now drafts — that documented paperwork share is what holds it at 14 instead of 18.
Hands-on in uncontrolled environments You are on the floor in a gym, pool, community bus, or day room, doing physical assists and gait-belt transfers with patients who fall, resist, or seize, in settings you do not control — a community outing to a grocery store or a fishing trip is the opposite of a fixed workstation, which puts this near the top rather than at the mid-range of jobs with 'some' physical work.
Certification preferred, not legally required CTRS credentialing through NCTRC is required by most hospitals and by CMS conditions of participation for inpatient rehab, but only a few states (NC, UT, NH, OK) license the title, so an unlicensed activity aide can legally do adjacent work in most of the country — that gap between employer-mandated certification and statutory practice protection is exactly a 10, not a nurse's 17.
Meaningful discretion You decide whether a patient is ready for an unsupervised community reintegration outing, whether a fall risk can attempt standing basketball, and when to report suicidal statements disclosed in a group — real calls with real consequences, but they are made inside an interdisciplinary team where the physician and case manager sign off on the plan and the discharge, which caps this at 12 rather than the 16 of the person who owns the decision alone.
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 (14/20 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 (10/20) is whether the law requires a licensed human to sign. Trust premium (17/20) 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 39 of this occupation's 69 points (57%).
Embodiment (16/20) 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 84/100, still SAFE.
Task-mix shift: this role has a genuine two-tier structure. If AI absorbs the plan drafting, progress notes, insurance justification and activity ideation, what remains is behavioral de-escalation, in-the-moment adaptation for cognitive/mobility limits, and family negotiation — none of which is documentable-away. The residual job is more resistant than the current average job.
Expansion of state practice acts for recreational therapy beyond the current handful (NC, UT, NH, OK licensure; NJ registration) — NCTRC has an active state-licensure advocacy agenda. A state law making 'recreational therapy' a protected scope with a licensed CTRS required to sign the treatment plan in the medical record would convert credential-preference into a legal signature requirement.
CMS Conditions of Participation for inpatient rehabilitation facilities and psychiatric hospitals (42 CFR 482.62) already name qualified therapeutic recreation staff; a tightening that names CTRS specifically as the required signer of the individualized activity/treatment plan, or a Joint Commission element of performance requiring a credentialed RT to attest that AI-drafted plans were reviewed, would raise the floor nationally.
Formal inclusion of the RT on documented risk-decision bodies — e.g., facility policy requiring RT sign-off on community-outing risk assessments, elopement risk, or least-restrictive-alternative determinations for psychiatric patients. Where RTs own the written 'is this patient safe to go off-unit' call, the accountability is explicit rather than informal.
Payer or accreditation recognition that group therapeutic recreation is a billable skilled service rather than bundled 'activities' — e.g., MDS/PDPM coding changes in skilled nursing that credit RT minutes toward reimbursement, which historically drove staffing levels. Buyers paying line-item for a human RT rather than absorbing it as overhead hardens the premium.
The limit. Embodiment is already near ceiling and cannot rise meaningfully. The main downside risk sits outside these levers: RT is often a discretionary cost center, and budget deletion — not AI substitution — is the realistic displacement path, which no liability or trust lever addresses.
| New York-Newark-Jersey City, NY-NJ | 1,090 | $66,050 +7% |
| Los Angeles-Long Beach-Anaheim, CA | 380 | $91,740 +48% |
| Washington-Arlington-Alexandria, DC-VA-MD-WV | 380 | $71,050 +15% |
| Philadelphia-Camden-Wilmington, PA-NJ-DE-MD | 360 | $58,970 -5% |
| Chicago-Naperville-Elgin, IL-IN | 340 | $54,460 -12% |
| Las Vegas-Henderson-North Las Vegas, NV | 340 | $78,730 +27% |
| Riverside-San Bernardino-Ontario, CA | 340 | $98,120 +58% |
| Atlanta-Sandy Springs-Roswell, GA | 260 | $60,640 -2% |
| Stockton-Lodi, CA | 50 | $115,330 +86% |
| Vallejo, CA | 50 | $115,330 +86% |
| Napa, CA | 150 | $107,810 +74% |
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 69. 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.