Menu
Social Worker
Three components - Automation Resistance, Structural Moat, and Demand - add up to 68.
This score covers the social-work family as a whole. Child/family/school, healthcare, mental-health/substance-use, and all-other roles differ in licensure, setting, and exposure; representative detailed rows are used because the family aggregate does not publish every AI/ORS/O*NET field.
Paperwork and resource search are exposed, especially notes, referrals, summaries, and eligibility checks across agencies, but trust, safety judgment, home and school context, mandated reporting, benefits navigation, and clinical responsibility keep direct replacement pressure limited.
Observed exposure is 0.74% for child/family/school social workers, 9.16% for healthcare social workers, and 0.00% for mental-health and substance-use social workers. Modeled median job-loss risk sits around 7% to 14% across those specialties. That supports strong resistance: interviews, trust, safety assessment, home or school context, mandated reporting, and licensed clinical lanes still need people.
AI can help with documentation, care-plan drafts, eligibility checks, referral search, translation, summaries, and risk-screening support. The worker upside is limited because public agencies, health systems, schools, and nonprofits can capture that efficiency through higher caseload expectations, compliance demands, or faster paperwork rather than higher pay.
Clinical licensure, accredited education, supervised hours, field settings, and low robotics risk create real protection, while the broader family still includes bachelor-level, school, child-welfare, and public-agency roles with thinner formal gates and lower autonomy for workers.
Federal physical data for child/family/school social workers shows a 5-pound median maximum lift, about 28% standing or walking time, and about 33% outdoor work. That is not heavy labor, but it is more embodied than pure desk work because social workers may enter homes, schools, hospitals, shelters, courts, and crisis settings.
Social work has accredited degree pathways, state licensure, supervised clinical hours, exams, and continuing education, especially for clinical practice. Federal requirements data shows license, certification, or registration required for about 37% of the representative child/family/school occupation. The protection is meaningful, but not uniform across every casework or community role.
Robotics is not the main threat to social work. The work is relational, cognitive, legal, clinical, and field-based: interviews, reporting decisions, family context, resource coordination, therapy, discharge planning, and safety judgment. Robots do not replace those responsibilities at scale; software pressure sits mostly in paperwork and triage.
The family mixes bachelor's-entry casework with master's-level and licensed clinical tracks. Child/family/school social work sits around a four-year preparation profile, while healthcare and mental-health social work often go deeper. The score uses the family center rather than treating every role as clinical private practice.
The social-work family is large and demand is real across healthcare, schools, child and family services, mental health, benefits systems, and public programs, but funding and caseload pressure hold the score below the strongest healthcare paths.
The social-work family is much larger than a narrow leftover category: about 810,900 jobs, about 74,000 annual openings, and roughly 5.5% projected growth. Annual openings are a little above 9% of the workforce, which supports a strong volume score for a public-service and healthcare-adjacent path.
Demand comes from aging, healthcare coordination, mental-health and substance-use needs, child and family services, schools, courts, benefits systems, and public replacement hiring. The evidence is broad and consistent, but the market is not purely private demand: budgets and reimbursement decide how much need becomes staffed roles.
Need is durable, local, regulated, and trust-based, but social work remains exposed to funding shocks, caseload stress, reimbursement limits, public budgets, and real-wage drag. A 2015 family median inflated into 2025 dollars is about $62,300, close to the 2025 family median of $61,780. The work is needed, but job quality can still be fragile.
The score would strengthen if public agencies, hospitals, schools, and insurers fund more staff, better supervision, and safer caseloads. Need alone is not the trigger; the change has to reach stable jobs, safer staffing, better pay, and better working conditions.
The score would weaken if agencies use AI mainly to process more cases with the same or fewer workers. Faster notes and triage would not be enough; the trigger is meaningfully higher risk per worker and less time with clients.
The score would strengthen if the roles readers most often pursue move toward licensed clinical social work with better pay, supervision, and autonomy. The gain has to be broad enough to describe the family, not only private-practice success stories or a few high-income metros.