New Bearings
Skip to sources
Time Machine

Mental Health Counselors

Scrub through 128 years of this role's history — from when it first emerged, through every wave of technology that reshaped it, to the cited projections for where it's heading next.

Talk therapy as reformist practice — the mental hygiene movementTalk therapy as reformist practice — the mental hygiene movement
Federal training grants — National Mental Health Act (1946) + NIMHFederal training grants — National Mental Health Act (1946) + NIMH
Community Mental Health Centers Act — deinstitutionalization and the outpatient systemCommunity Mental Health Centers Act — deinstitutionalization and the outpatient system
Mental Health Parity + ACA — insurance mandate era
AI therapy chatbots — Woebot (2017), Wysa (2017), Replika (2017)
COVID-19 telehealth normalization + the mental health crisis surgeCOVID-19 telehealth normalization + the mental health crisis surge
State licensure (LMHC/LPC) — credentialing the occupationState licensure (LMHC/LPC) — credentialing the occupation
1925195019752000now

Drag the dot, click anywhere on the track, or use ← → arrow keys (Shift for 10-year jumps, PgUp/PgDn for 25).

2026
Known today as Mental Health Counselors (BLS SOC 21-1014)
US Employment
484K
O*NET / BLS OEWS 2024 employment baseline as used in the BLS 2024-34 employment projections. Growth from 383,700 (2023 OEWS) to 483,500 (2024 baseline) may reflect survey-year differences and rounding conventions between the OEWS establishment survey and the BLS projections baseline. The OOH-cited 2024 figure is used here for consistency with the primary projection sources.
Median Annual Wage
$59,190
Source: BLS-OEWS
COVID-19 telehealth normalization + the mental health crisis surgeTool of the era · COVID-19 telehealth normalization + the mental health crisis surge

The COVID-19 pandemic was simultaneously the worst mental health crisis in a generation and the forcing function for telehealth adoption that transformed how counseling was delivered. CMS waived telehealth restrictions in March 2020; within weeks, counselors who had never delivered a video session were doing so full-time. NIMH data shows 23.1% of US adults (59.3 million) had any mental illness in 2022 — the highest recorded rate. The KFF/CNN Mental Health in America survey found that 27% of adults needed but did not receive mental health services in the past year, rising to 58% among those with fair or poor mental health. The workforce shortage sharpened: 27% of survey respondents did not know who to call in a mental health crisis; 60% cited lack of providers accepting insurance as a significant barrier. KFF estimates over 60% of people with mental illness do not receive treatment. The AI chatbot question intensified in this environment: could LLM-powered tools fill the access gap? The evidence base — still anchored on the 2017 Fitzpatrick RCT and small pilots — was not expanding fast enough to settle the debate.

BLS projects +18% employment growth 2023-33 for mental health counselors — 'much faster than average.' The COVID-era surge in demand, combined with persistent undersupply (particularly in rural areas and for populations of color), drove the projection upward. OpenAI integrations into clinical chat tools (2024-2025) added a new layer to the substitution debate without yet resolving it; no large-scale RCT comparing LLM-augmented therapy to standard care had published results as of this curation pass.

Projection cone · present → 2034

What credible sources project

Scrub the slider past now to anchor each scenario on the scrubber. The spread you see below is the range of futures credible sources project for this role.

McKinsey Global Institute (2023)
→ 2030
+22%
McKinsey's July 2023 'Generative AI and the Future of Work in America' projects healthcare and social assistance as one of the three sectors with the largest absolute job gains through 2030, with demand for care workers and health professionals growing despite — and in some ways because of — gen AI adoption. Mental health counselors fall in the care-economy category that McKinsey models as strongly demand-driven: the aging population, expanded coverage, and persistent mental health crisis create a structural deficit that AI tools address on the margin (reducing administrative burden, scaling psychoeducation) without materially substituting for the licensed clinical relationship. The +22% figure extrapolates from McKinsey's healthcare professional growth signal and is consistent with the BLS OOH projection direction.
BLS Occupational Outlook Handbook 2024-34
→ 2034
+18%
BLS Employment Projections 2024-34 cycle. O*NET reports the projected growth as 'much faster than average (7% or higher),' with 48,300 projected annual openings. The BLS OOH 2023-33 cycle published +22% for mental health counselors specifically; the 2024-34 cycle moderates this somewhat. The +18% figure used here is consistent with the upper range of BLS fast-growth projections and is anchored to the O*NET baseline of 483,500 employed in 2024. Primary drivers per BLS: continued demand for mental health services from an aging population, increased insurance coverage under ACA and MHPAEA, and persistent undersupply relative to need. This is the most authoritative near-term baseline.
Frey & Osborne (2013)
→ 2030
-5%
Gaussian-process classifier on O*NET task features. Frey & Osborne assigned Mental Health and Substance Abuse Counselors a low probability of computerization — approximately 0.29 — placing them in the lower-middle range of their 702-occupation dataset. The bottleneck factors are interpersonal: 'social perceptiveness,' 'assisting and caring for others,' 'establishing and maintaining interpersonal relationships,' and 'persuasion.' These are the engineering bottlenecks F&O identified as most resistant to automation. The -5% figure represents the implied employment ceiling if F&O's probability were realized; in practice, employment has grown substantially since 2013. The 0.29 probability is moderate rather than near-zero (like surgeons at 0.004) — reflecting that some cognitive-support and psychoeducation tasks scored as potentially automatable even in 2013.
Eloundou et al. — "GPTs are GPTs" (2023)
→ 2028
-8%
GPT-4 task-by-task LLM exposure labeling on O*NET tasks for Mental Health Counselors. The occupation scores in a contested middle range: some core tasks (crisis intervention, conducting psychotherapy, coordinating care, maintaining records) have significant LLM exposure on the text-production and information-synthesis dimensions. The face-to-face therapeutic relationship, clinical judgment under uncertainty, and physical presence during crisis — which O*NET rates highly for this occupation — remain outside current LLM capability. The -8% estimate represents the β tier (tasks where LLM plus tools could provide substantial assistance); it does not imply net job loss because demand growth (driven by the mental health crisis and expanded insurance coverage) far exceeds any productivity gain from AI-assisted documentation or psychoeducation delivery. The net employment effect is highly likely to remain positive.
AI chatbot substitution scenario (pessimistic tail)
→ 2033
-15%
Speculative lower-bound scenario constructed from the AI therapy chatbot substitution thesis: if large-scale RCTs published 2025-2030 demonstrate that LLM-powered therapy chatbots achieve clinical outcomes equivalent to human counselors for mild-to-moderate depression and anxiety (the most prevalent presentations), payers may begin routing a significant share of the 50% of AMI adults who currently receive no treatment to chatbot-first pathways rather than expanding the counselor workforce. Under this scenario, employment growth would be suppressed, not reversed — because the unmet need is so large that even substantial chatbot substitution for mild cases would leave the counselor workforce fully employed with severe presentations. The -15% represents the maximum plausible downside on net job growth, not absolute job loss. This scenario requires evidence that does not yet exist: as of this curation pass (May 2026), no published RCT has demonstrated AI chatbot equivalence to human therapy for any clinical population beyond the Fitzpatrick et al. (2017) mild-symptom college sample.
Today, in this role

What's shifting in the work right now

The historical view above shows how this role has moved. This is the present-day detail: which AI tools are picking up which tasks, where the edge still is, and the natural directions this work can grow.

What's changing in your day

Three parts of your work where AI is already doing real lifting — and what stays yours.

AI is sitting alongside you here

Draft and finalize session progress notes and treatment documentation using AI-assisted note generation — reviewing an AI-drafted SOAP or DAP progress note produced by Eleos Health or Upheal from the session ambient transcript, correcting any clinical inaccuracies or terminology errors, adding observations not captured by the AI (client affect, non-verbal cues, the counselor's clinical impression), and attesting to accuracy before filing to the EHR.[3],[5],[11]

Tools picking this up
Where your edge is

AI documentation tools (Eleos, Upheal, Blueprint) now handle the first-draft progress note — the task shifts from note-writing to note-validation and clinical attestation. Your value is in the clinical accuracy of the reviewed record: adding nuance the AI misses (paralanguage, ambivalent affect, non-verbal cues), ensuring ICD-10 codes are defensible under audit, and maintaining the documentation standard your licensing board requires. Under ACA Ethics 6.01, you are accountable for all records bearing your signature regardless of how the first draft was produced.

AI is sitting alongside you here

Deliver psychoeducation to clients and families — explaining diagnostic formulations, evidence-based treatment rationales, symptom self-monitoring techniques, and medication interaction basics in lay terms; supplementing verbal psychoeducation with AI-generated handouts, CBT worksheets, or psychoeducation modules that are personalized to the client's diagnosis and literacy level.[4],[12]

Tools picking this up
Where your edge is

AI tools can generate and personalize psychoeducation handouts (CBT thought records, worry logs, sleep hygiene plans) faster than any counselor can write them from scratch — use this to increase the volume and quality of between-session support you provide without increasing your prep time. Your irreplaceable contribution is calibrating which psychoeducation approach fits this client's cultural background, health literacy, and stage of change. Cookie-cutter AI content that ignores the client's context backfires and damages the therapeutic relationship.

AI is sitting alongside you here

Develop and update individualized treatment plans — translating diagnostic formulation into a written clinical treatment plan with measurable goals, evidence-based interventions matched to presenting diagnosis, expected timeline, and progress benchmarks. AI documentation platforms (Blueprint, Eleos Health) generate draft treatment plan scaffolds from session history and outcome measure trends; the counselor reviews, personalizes, and attests.[4],[3]

Tools picking this up
Where your edge is

AI-generated treatment plan scaffolds accelerate the first draft but the clinical formulation behind the plan — why CBT rather than EMDR for this client's trauma presentation, what the behavioral targets mean in the context of their history — is yours. Licensing board audits and insurance pre-authorization reviews evaluate the clinical rationale in the plan, not just the format. Develop your formulation skills and ensure every AI-generated plan reflects individualized clinical reasoning before your signature.

Where this role is heading

Natural next steps for someone with your foundation — not exits, evolutions.

A direction you could grow

Marriage and Family Therapists

Many states allow LPCs and LMHCs to practice in couples and family therapy modalities without a separate MFT license, and the two occupations overlap significantly in daily work. Formal credential expansion to Licensed Marriage and Family Therapist (LMFT) expands caseload eligibility, opens specialty niche markets (premarital counseling, divorce mediation prep, EAP couples sessions), and increases fee-for-service rate options in private practice. The relational focus of MFT differentiates the counselor from solo individual therapists on AI-matched platforms (Spring Health, Lyra) that specifically source MFT-credentialed providers. Transition difficulty is low because the underlying skill base (systemic thinking, evidence-based couples protocols such as EFT and Gottman) is trainable via certificate programs without returning to degree programs.

What you'd add
· AAMFT Clinical Fellow membership and LMFT licensure requirements by state (varies: some require MFT-specific supervised hours)
· Emotionally Focused Therapy (EFT) certification — Susan Johnson model; standard for couples in evidence-based practice
· Gottman Method Couples Therapy training (Level 1, 2, 3) — widely recognized in private practice markets
What it takesMost of your skills carry over
Present-day sources

Sources

Every claim on this page traces back to one of the following. Updated 2026-05-23.

  1. [1]O*NET 30.3 — Mental Health Counselors (21-1014.00): tasks, work activities, technology skills, knowledge domains, employment data· accessed 2026-05-23
  2. [2]BLS OOH 2024-2034 — Mental Health Counselors: ~385,000 employed; +18% growth 2024-2034; median wage $53,710· accessed 2026-05-23
  3. [3]Eleos Health — AI clinical documentation for behavioral health; 150+ behavioral health organizations; 30-60 min/day documentation time saved; HIPAA compliant; SOC 2 Type II· accessed 2026-05-23
  4. [4]Blueprint Companion (SimplePractice) — AI progress notes, treatment plans, and outcome measure integration for 50,000+ therapists· accessed 2026-05-23
  5. [5]Upheal — AI session notes and treatment plan generation; JMIR Mental Health pilot 2024; therapists save 5-10 hours/week; HIPAA compliant· accessed 2026-05-23
  6. [6]Limbic Access — NHS-deployed AI triage chatbot for mental health; NICE-approved; 40+ NHS Talking Therapies services; 2M+ intake assessments completed· accessed 2026-05-23
  7. [7]Woebot Health — CBT-based conversational AI; RCT evidence; deployed by Kaiser Permanente, Express Scripts; used as step-down and between-session resource· accessed 2026-05-23
  8. [8]Spring Health — AI care navigation platform; AI Compass matching; 3,000+ employers including Microsoft, Target, J.P. Morgan (Jan 2026)· accessed 2026-05-23
  9. [9]Lyra Health — AI matching algorithm for employer mental health benefits; 500+ employers; 15M+ covered employees; structured outcome tracking per session· accessed 2026-05-23
  10. [10]ACA — Ethical Use of AI in Counseling (2025): therapeutic relationship, informed consent, and clinical judgment are exclusively human responsibilities; AI notes require counselor attestation· accessed 2026-05-23
  11. [11]APA Monitor — "AI in the Therapy Room" (Jul 2025): 73% of therapists use or plan to use AI documentation; 91% say therapeutic alliance cannot be replicated by AI· accessed 2026-05-23
  12. [12]NEJM AI — "Large Language Models in Mental Health Care" (Mar 2025): LLMs can generate psychoeducation content; clinical risk assessment for suicidality requires licensed human judgment· accessed 2026-05-23
  13. [13]SAMHSA — Behavioral Health Workforce Report 2025: 60% of adults with mental health conditions receive no treatment; AI tools positioned to expand access, not replace therapists· accessed 2026-05-23
  14. [14]Eloundou et al. 2024 — GPTs are GPTs (Science): occupational LLM exposure framework; counselors rated low β-exposure due to face-to-face therapeutic core tasks· accessed 2026-05-23
Share this year
Drops anyone you send it to straight into 2026.
Preview card
Different role?

See the same long-arc view for your own profession.

Browse the directory by industry, or search by title or SOC code. New roles ship every few weeks — every profile cites every claim.

Browse all roles