SAFE
SLP work is hands-on and relational: positioning a stroke patient for a swallow study, cueing articulation in real time, coaxing a nonverbal three-year-old through play-based therapy, deciding whether a dysphagia patient is safe to eat by mouth. AI can draft evaluation reports, score standardized tests, transcribe sessions, and generate home practice materials — the documentation load, which is real, but not the treatment. State licensure plus CCC-SLP credentialing and personal accountability for aspiration and feeding decisions keep a human clinician on the chart.
Dipped in 2020, then grew past where it started.
Median pay $79,120 → $97,870 -1.0% 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
+15% 187,400 → 215,500 on the projections basis
Hard to automate, and growing
The work resists current AI and the BLS projects +15% 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.
~13,300 openings a year on average, including replacing people who leave.
Oral TherapistSpeech ClinicianSpeech TherapistVoice PathologistSpeech PathologistLanguage PathologistSpeech-Language SpecialistSpeech and Language TeacherSpeech Language PathologistSpeech Correction ConsultantSpeech and Language ClinicianSpeech and Language TherapistPublic School Speech ClinicianPublic School Speech TherapistSpeech and Language SpecialistSpeech-Language Pathologist (SLP)School SLP (School Speech Language Pathologist)Travel SLP (Travel Speech Language Pathologist)SNF RN (Skilled Nursing Facility Registered Nurse)SLP CF (Speech Language Pathologist Clinical Fellow)Bilingual Speech-Language Pathologist (Bilingual SLP)Pediatric SLP (Pediatric Speech Language Pathologist)Pediatric Speech-Language Pathologist (Pediatric SLP)Home Health SLP (Home Health Speech Language Pathologist)
Holding it up: trust premium . Weakest point: embodiment .
Tasks largely resist digitisation The minute-to-minute work is contingent responding — you hear a distorted /r/, decide whether to cue placement or back off to a syllable level, and adjust within the same breath; standardized test scoring, IEP goal templating, and progress-note drafting are genuinely automatable, which is what keeps this at 16 rather than 19.
Hands-on in uncontrolled environments You are placing your hand under a child's jaw for oral-motor cueing, positioning a post-CVA patient upright at 90 degrees for a bedside swallow screen, suctioning a trach patient during a Passy-Muir trial, and doing it in NICUs, nursing-home dining rooms and elementary classrooms — physical and uncontrolled, though you're not lifting or in hazard gear, which is why this sits at 15 and not 18.
Licensed human required and personally liable State licensure is mandatory in all 50 states and the ASHA CCC-SLP plus Medicare Part B billing requires your NPI on the plan of care, so when a patient you cleared for a Level 2 dysphagia diet aspirates, the incident review names you — a 16 rather than a physician's 19 because you work under physician referral for medical dysphagia and don't hold independent prescriptive authority.
Exists to be accountable for ambiguous calls You decide whether a MBSS finding means NPO, whether a stutter is developmental or warrants a fluency diagnosis at age four, whether a nonspeaking child needs AAC now versus more time on verbal targets, and whether to discharge — calls made on incomplete data with aspiration pneumonia or years of lost communication access on the other side.
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 (16/20 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 (16/20) is whether the law requires a licensed human to sign. Trust premium (18/20) is whether buyers specifically pay for a person. Judgment and accountability (18/20) 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 52 of this occupation's 83 points (63%).
Embodiment (15/20) 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 92/100, still SAFE.
Malpractice carriers or hospital credentialing bodies requiring named-clinician attestation on any AI-assisted evaluation report or aspiration-risk determination, mirroring radiology AI attestation requirements now appearing in insurer policies
Task-mix shift: if AI absorbs the routine tier (standardized test scoring, IEP/eval report drafting, session notes, home-program generation, Medicare productivity documentation), the residual job is disproportionately instrumental swallow interpretation, AAC device candidacy decisions, tracheostomy/ventilator patients, and behaviorally complex pediatric cases — all judgment tier. This raises task_resistance without any new law
CMS conditions of participation or state licensure rules explicitly requiring a licensed SLP (not an aide, not a software output) to personally sign dysphagia diet-level recommendations and instrumental swallow study interpretations (MBSS/FEES), plus state boards restricting delegation of swallowing evaluation to SLPAs — ASHA already lobbies on SLPA scope, and several states (e.g., Texas, Florida) have codified SLPA supervision limits
Formal designation of the SLP as the accountable clinician on interdisciplinary dysphagia and airway teams (NPO decisions, PEG-tube recommendations), and IDEA due-process rulings that hold the evaluating SLP — not the district or its software — accountable for eligibility determinations
Little headroom: buyers are already overwhelmingly payers (Medicare, Medicaid, school districts) who purchase the licensed credential rather than a specific person. The only realistic route is growth of private-pay pediatric and accent/voice practices where parents select a named clinician, plus school-district contracts that bar teletherapy-vendor substitution — some state legislatures (e.g., teletherapy caps in school SLP contracts) have moved this way
The limit. Already at 83; the realistic ceiling is high-80s. The main downside risk is not AI replacing treatment but payers substituting cheaper labor tiers — SLPAs, teletherapy platforms with high caseload ratios — under an AI-assisted supervision model. That would erode trust_premium and liability_shield even as task_resistance holds.
| New York-Newark-Jersey City, NY-NJ | 15,260 | $107,860 +10% |
| Chicago-Naperville-Elgin, IL-IN | 7,150 | $98,490 +1% |
| Dallas-Fort Worth-Arlington, TX | 4,800 | $99,540 +2% |
| Los Angeles-Long Beach-Anaheim, CA | 4,800 | $109,740 +12% |
| Houston-Pasadena-The Woodlands, TX | 3,780 | $100,400 +3% |
| Boston-Cambridge-Newton, MA-NH | 3,720 | $102,810 +5% |
| Philadelphia-Camden-Wilmington, PA-NJ-DE-MD | 3,330 | $100,080 +2% |
| Washington-Arlington-Alexandria, DC-VA-MD-WV | 3,140 | $99,960 +2% |
| El Centro, CA | 70 | $133,760 +37% |
| San Jose-Sunnyvale-Santa Clara, CA | 890 | $132,720 +36% |
| Boulder, CO | 320 | $131,620 +34% |
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 83. 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.