Exercise Trainers and Group Fitness Instructors
39-9031.00Median wage $47,160/yr322,930 employed (US)Rank #537 of 923 scored · top 58% by substitution
Instruct or coach groups or individuals in exercise activities for the primary purpose of personal fitness. Demonstrate techniques and form, observe participants, and explain to them corrective measures necessary to improve their skills. Develop and implement individualized approaches to exercise.
Sub-scores
0–100 · band = confidence interval from rater disagreement
Substitution26
Exposure18
Augmentation56
Substitution — the headline: capability discounted by cost, barriers and adoption.
Exposure — technical capability alone, regardless of whether anyone deploys it.
Augmentation — how much AI assists without replacing. High here + moderate substitution = a changing job, not a disappearing one.
Tasks on the substitution scale
20 rated tasks, binned by substitution score.
Position among all scored occupations
Distribution of 923 occupation scores; the marker is this occupation.
Tasks with substitution ≥ 70
0%
Run 1.0.0-draft.1 · computed 2026-08-05 · rater panel: claude-sonnet-5, claude-haiku-4-5-20251001 · intervals span rater disagreement.
Why this score
The five weighted dimensions of the composite, averaged across this occupation's tasks (importance-weighted, panel mean). Exact weights and formulas: /api/v1/methodology.
Task automatabilityw 35%17
panel mean rating 1.7/5 → substitution pressure 17/100
Technical feasibility todayw 20%20
panel mean rating 1.8/5 → substitution pressure 20/100
Cost vs. human wagew 15%26
panel mean rating 2.0/5 → substitution pressure 26/100
Adoption barriersw 20%inverted — strong barriers lower the score47
panel mean rating 3.1/5 (barrier strength) → substitution pressure 47/100
Sector adoption velocityw 10%23
panel mean rating 1.9/5 → substitution pressure 23/100
Task breakdown (20 tasks)
Substitution pressure per task, weighted by O*NET importance in the composite. Expand a task for the full rater audit trail — every rating, every model, every rationale.
Promote health clubs through membership sales, and record member information.
61CI 55–66 · exposure 50 · augmentation 75 · importance 3.1/5 · click for rater detail
Promote health clubs through membership sales, and record member information.
61| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 4/5 | Health clubs and fitness chains have been early adopters of CRM and chatbot technology for membership sales and data management, with many mid to large-sized facilities already using automated systems and AI-assisted prospecting in production. |
| Sector adoption velocity | claude-sonnet-5 | 3/5 | Fitness/retail sector has moderate digitization with CRM and marketing automation tools increasingly common, but personal, in-person sales culture in small gyms slows full displacement. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI tools substantially assist fitness instructors and sales staff by automating follow-up emails, lead qualification, and data logging, allowing staff to focus on relationship-building and closing high-intent prospects, thereby raising overall sales productivity. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI-driven CRM, lead scoring, automated follow-up emails, and record management meaningfully boost trainer/staff productivity in the sales and admin process. |
| Task automatability | claude-haiku-4-5-20251001 | 3/5 | The membership sales and information recording portions of this task are partially automatable—AI chatbots and CRM systems can handle initial inquiries, membership applications, and basic data entry. However, closing sales typically requires human relationship-building and persuasion, preventing full automation at the ≥50% time-saving threshold. |
| Task automatability | claude-sonnet-5 | 3/5 | CRM data entry and templated marketing outreach can be substantially automated, but sales conversion and relationship-building still benefit from human persuasion, so only part of the task meets the 50% bar off-the-shelf. |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | Few regulatory or licensing barriers exist for automating membership sales and record-keeping in fitness contexts. However, members may prefer human interaction, and some organizational friction around replacing sales staff persists. |
| Adoption barriers | claude-sonnet-5 | 1/5 | No licensing or regulatory requirement exists for membership sales or record-keeping; it's a low-barrier commercial task. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 4/5 | AI-based membership sales and CRM tools are relatively inexpensive to operate at scale (per-transaction costs are low), and the information recording is largely automatable, making AI significantly cheaper than hiring dedicated sales staff for this specific function. |
| Cost vs. human wage | claude-sonnet-5 | 3/5 | Automated CRM and marketing tools are cheap relative to a trainer's time for record-keeping, but sales conversion work still requires human involvement, keeping overall cost roughly comparable when blended. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 3/5 | Deployed products exist for membership management and sales assistance (CRM platforms with chatbots, automated form handling), but they struggle with complex objection handling and personalized sales pitches that drive actual conversions. Most require significant human oversight. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | CRM platforms and marketing automation tools (email campaigns, chatbots for lead qualification, automated data capture) are deployed widely in gyms, but actual closing of sales and personalized promotion still often involves staff. |
Provide students with information and resources regarding nutrition, weight control, and lifestyle issues.
49CI 40–59 · exposure 42 · augmentation 75 · importance 3.8/5 · click for rater detail
Provide students with information and resources regarding nutrition, weight control, and lifestyle issues.
49| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 3/5 | Fitness is moderately digitized (apps, online coaching platforms), and some chains use AI nutrition tools, but the sector remains highly dependent on in-person instruction and human relationships, limiting deep automation of this task despite growing interest in digital supplements. |
| Sector adoption velocity | claude-sonnet-5 | 3/5 | Fitness and wellness apps have adopted AI-driven content and chatbots moderately, but the broader personal training industry remains largely in-person and slow to formally integrate AI tools. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI can substantially assist instructors by generating customized nutrition plans, suggesting resources, tracking client progress, and answering routine questions, allowing trainers to focus on motivation, form correction, and relationship-building rather than content creation. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI can efficiently draft nutrition guides, meal plans, and lifestyle tips for trainers to customize and deliver, significantly speeding up content preparation while the trainer maintains the relationship. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI can generate basic nutrition and lifestyle information, delivering personalized guidance requires understanding individual health contexts, medical histories, and motivational cues that demand human judgment. The task involves real-time feedback and relationship-building that current systems cannot replicate end-to-end with quality parity. |
| Task automatability | claude-sonnet-5 | 3/5 | AI chatbots can generate generic nutrition and lifestyle information and personalized plans based on user input, covering much of the informational content, but cannot replace in-person accountability, motivation, and adaptive coaching relationships. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Fitness professionals often require certifications (ACE, NASM, etc.) that include nutrition education standards, and liability concerns around medical advice create friction; customers also expect human relationship and accountability from their instructor, not a bot delivering generic advice. |
| Adoption barriers | claude-sonnet-5 | 2/5 | Nutrition advice from trainers is often informal and not tightly regulated like clinical dietetics, though some liability concerns and scope-of-practice norms exist for specific medical conditions. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 4/5 | AI-generated nutrition and lifestyle content is very cheap to produce and deliver at scale; however, oversight and integration into instructor workflows add modest cost, keeping the ratio favorable compared to paying a human nutritionist or health coach. |
| Cost vs. human wage | claude-sonnet-5 | 4/5 | AI-generated nutrition content and generic advice cost pennies compared to a trainer's hourly rate, though quality assurance and personalization add some overhead. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 3/5 | Chatbots and AI tools can provide generic nutrition facts and lifestyle resources, and some fitness apps incorporate AI-driven nutrition guidance, but deployed systems lack the personalization, credibility, and adaptive coaching necessary for reliable real-world performance in group or individual settings. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | Consumer apps (MyFitnessPal, various AI nutrition coaches, ChatGPT) already provide diet and lifestyle guidance at scale, but accuracy varies and they lack certification or liability coverage for tailored medical/nutrition advice. |
Advise clients about proper clothing and shoes.
48CI 23–74 · exposure 38 · augmentation 63 · importance 3.3/5 · click for rater detail
Advise clients about proper clothing and shoes.
48| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Fitness facilities are digitizing but slowly; most trainers still rely on face-to-face interactions and informal advice; AI tools for this specific advisory task have minimal market penetration. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Personal fitness training is a low-digitization, high-touch, small-business-dominated sector where AI adoption for interpersonal coaching tasks remains limited and pilot-stage at best. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI could assist trainers by suggesting evidence-based footwear categories or clothing recommendations based on activity type, which the trainer then personalizes and validates with the client. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI can effectively supplement a trainer by providing quick reference info, product suggestions, or personalized shoe-fit guidance, enhancing but not replacing the client relationship. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | Advising clients on clothing and shoes requires contextual understanding of individual body types, fitness goals, injury history, and personal preferences—nuanced judgment that current AI systems cannot reliably perform end-to-end without substantial human oversight. |
| Task automatability | claude-sonnet-5 | 4/5 | Giving generic clothing/footwear advice is simple factual/informational guidance that chatbots can already produce accurately and quickly, saving most of the verbal explanation time.But a small residual of in-person, client-specific assessment (fit, gait, foot shape) may not be fully replaceable. |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | Clients generally expect personalized, in-person advice from their trainer, and liability concerns around foot support or injury prevention create some friction, though no strict legal requirement prevents AI assistance. |
| Adoption barriers | claude-sonnet-5 | 1/5 | No licensing, liability, or regulatory requirement mandates a human deliver clothing/shoe advice; it's informal guidance with minimal risk. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | The cost of AI systems capable of personalizing clothing and shoe advice (including computer vision for fit assessment and training) would exceed the marginal time a trainer spends on this brief advisory task. |
| Cost vs. human wage | claude-sonnet-5 | 5/5 | Text-based advice on clothing/shoes is trivial for an LLM to generate, costing fractions of a cent versus a trainer's per-session wage for this sub-task. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | While AI can provide generic advice on footwear types or clothing guidelines via chatbots, no deployed product reliably advises individual clients on proper fit and suitability in real fitness settings; this remains largely human-driven. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | General fitness apps and chatbots can give clothing/shoe recommendations today, but no deployed product specifically integrates this into a trainer's live coaching workflow reliably. |
Maintain equipment inventories, and select, store, or issue equipment as needed.
41CI 35–47 · exposure 30 · augmentation 63 · importance 3.7/5 · click for rater detail
Maintain equipment inventories, and select, store, or issue equipment as needed.
41| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 3/5 | Fitness facilities show moderate adoption of inventory management software and digital tracking systems, but most adoption remains in larger commercial gyms and franchises. Smaller independent facilities lag significantly in digitization of this function. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Fitness and gym operations are a low-digitization, physically-oriented sector with slow uptake of AI-driven inventory automation compared to information-sector norms. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | Inventory management systems effectively assist staff by automating tracking, generating reports, and suggesting reorder points, significantly reducing manual counting and search time while the human maintains oversight of physical storage and condition checks. |
| Augmentation potential | claude-sonnet-5 | 3/5 | Inventory management apps and barcode/RFID systems can meaningfully assist trainers in tracking and reordering equipment, though physical handling remains manual. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While inventory tracking and issuance can be partially automated with RFID or barcode systems, the physical selection, storage optimization, and condition assessment of fitness equipment requires human judgment and hands-on handling. Current AI cannot manage the spatial and mechanical aspects end-to-end. |
| Task automatability | claude-sonnet-5 | 2/5 | Tracking and issuing physical gym equipment requires physical handling and on-site presence that current AI cannot perform end-to-end; only the inventory record-keeping portion is automatable.','confidence':0}, |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | There are minimal legal or regulatory barriers to automating inventory tracking, though some facilities may prefer human verification for quality assurance and insurance purposes. No licensing requirement protects human involvement in this task. |
| Adoption barriers | claude-sonnet-5 | 1/5 | No licensing or regulatory requirement mandates a human for equipment inventory or issuance; it's a low-stakes operational task. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Inventory management software is relatively affordable, but the ongoing human labor for physical equipment handling, inspection, and storage organization remains substantial. The AI system cannot eliminate most of the human cost for this task. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | Digital inventory tools are cheap, but the physical tasks still require paid staff time, so overall cost savings versus a human doing the full task are limited. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 3/5 | Inventory management software and asset tracking systems exist in production at fitness facilities, but they require significant human input for physical handling and condition assessment. Most deployments are hybrid, with AI handling database tracking while humans execute selection and storage decisions. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Inventory management software exists and is widely deployed, but it only covers the digital tracking aspect, not physical selection, storage, or issuing of equipment. |
Observe participants and inform them of corrective measures necessary for skill improvement.
34CI 30–39 · exposure 25 · augmentation 63 · importance 4.9/5 · click for rater detail
Observe participants and inform them of corrective measures necessary for skill improvement.
34| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | The fitness industry remains highly dependent on in-person, real-time human interaction; while wearables and fitness apps are widely used, they complement rather than displace trainers, and adoption of AI-driven form feedback remains experimental and limited to early-adopter boutique studios. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Fitness industry, especially small studios and personal training, has been slow to adopt AI-driven coaching compared to digitized white-collar sectors, though some big gyms and apps have begun piloting AI form feedback. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI-powered video analysis, real-time pose estimation overlays, and rep-counting assist trainers by freeing them to focus on motivation, breathing cues, and client psychology while the system flags gross deviations, substantially raising trainer productivity in group settings without the trainer leaving the loop. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI-powered wearables and camera-based apps can supplement trainers by giving quantitative movement data or flagging deviations, offering moderate assistance while the trainer still delivers hands-on judgment and corrections. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI vision systems can detect gross postural deviations and compare them to reference models, they cannot reliably perceive the nuanced, individualized cues (effort, breathing, fatigue, psychological readiness) that trainers use to calibrate corrections. End-to-end automation would require both reliable real-time biomechanical analysis and personalized coaching judgment, neither of which current systems achieve at scale without significant human oversight. |
| Task automatability | claude-sonnet-5 | 2/5 | This requires real-time physical observation of body mechanics, form, and movement quality in a live setting, which current AI systems (even with computer vision) cannot reliably replace end-to-end at equal quality for general fitness settings today.' |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | There are no hard legal requirements for a licensed human (fitness certification is often voluntary), but strong soft barriers exist: participants expect live human feedback, coaching relationships build retention and compliance, and liability for injuries from incorrect AI-generated form advice creates organizational hesitation and insurance complications. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing strictly requires a human for general fitness instruction, but liability concerns around injury from incorrect form and client preference for hands-on human coaching create moderate friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | A trainer's loaded wage (salary + benefits + facility overhead) is typically $20–40/hour in group settings; building and maintaining custom computer vision pipelines, integrating real-time feedback systems, and providing oversight still costs $10–20/hour per participant when amortized, making the ratio unfavorable for displacement at scale. |
| Cost vs. human wage | claude-sonnet-5 | 3/5 | App-based AI form-checkers are cheap per use, but they don't fully replace the trainer's task, so a fair cost comparison for the full task (including personalized real-time correction) is roughly comparable once oversight and limitations are factored in. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Computer vision products exist for form analysis and some fitness apps offer automated rep counting and form feedback, but these systems have high error rates in varied lighting, clothing, and body types; they also lack the contextual judgment to know *when* and *how* to correct based on individual capability and goals. Deployed systems are typically narrow demonstrations, not production-ready replacements for live instruction. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Some pose-estimation apps (e.g., AI form-checking apps) exist and give feedback on specific exercises, but they are narrow in scope, error-prone with varied body types/angles, and not deployed as a full substitute for a live trainer's corrective coaching. |
Plan physical education programs to promote development of participants' physical attributes and social skills.
34CI 30–39 · exposure 25 · augmentation 75 · importance 3.8/5 · click for rater detail
Plan physical education programs to promote development of participants' physical attributes and social skills.
34| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Fitness facilities are adopting scheduling and tracking tools but remain heavily reliant on human trainers for core delivery; automation in this sector lags because physical presence, motivation, and behavioral assessment remain high-value human functions. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Fitness and physical education sectors, especially in schools and community settings, show low-to-moderate AI adoption compared to fast-digitizing professional services sectors. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI can meaningfully assist trainers by generating evidence-based program templates, tracking participant progress, suggesting exercise modifications, and identifying gaps in physical or social development—allowing trainers to focus on relationship and real-time adaptation. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI can meaningfully assist trainers by generating program ideas, tracking progress data, and suggesting activity variations, while the instructor still designs and delivers the actual sessions. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | AI can assist with generating program templates, exercise databases, and participant tracking, but cannot independently assess individual physical attributes, social dynamics, or adjust programs in real-time based on live observation and participant feedback—core requirements for effective program planning. |
| Task automatability | claude-sonnet-5 | 2/5 | AI can help draft generic program templates, but tailoring plans to specific groups' physical development and social dynamics requires ongoing observation, adjustment, and human judgment that current systems cannot fully replace. |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | Gyms and fitness organizations face moderate friction from client expectations for human interaction and trust, insurance/liability concerns around program safety, and the nuanced social-skills component that clients expect from a human instructor. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No strict licensing mandates AI cannot be used for planning, but institutional and parental expectations around qualified instructors overseeing youth development create moderate friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | AI tools for program planning are relatively inexpensive, but a trainer's total cost includes real-time adjustment, motivation, safety supervision, and relationship-building that AI cannot replace; the combined human+AI cost often exceeds hiring a trainer directly. |
| Cost vs. human wage | claude-sonnet-5 | 3/5 | AI-assisted templates can cut planning time cheaply, but the need for human customization, in-person delivery, and revision keeps overall costs roughly comparable to a trainer doing it directly. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | While AI tools exist for workout generation and scheduling, no deployed product reliably performs the full task of personalized program planning that meaningfully develops both physical and social skills without significant human oversight and modification. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Some fitness apps and AI tools generate workout plans, but no deployed product reliably plans comprehensive physical education programs targeting both physical and social skill development at scale. |
Monitor participants' progress and adapt programs as needed.
32CI 25–39 · exposure 25 · augmentation 75 · importance 4.8/5 · click for rater detail
Monitor participants' progress and adapt programs as needed.
32| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Fitness tracking and app-based guidance (Peloton, Apple Fitness+) have grown, but group fitness still relies on live instructors; most adoption is in supplementary digital tools rather than AI-driven program adaptation replacing human trainers. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Fitness industry adoption of AI is growing in consumer apps but group fitness and personal training remain largely human-delivered, physical, and slow to digitize fully. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI-powered wearables and dashboards can help trainers visualize participant effort, form cues, and progress trends in real time, enabling faster and more targeted feedback and program adjustments during and after sessions. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI tools significantly help trainers track metrics, flag trends, and suggest program tweaks, letting the human focus on hands-on coaching and motivation while data-informed decisions improve outcomes. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | AI can track quantitative metrics (reps, weight, heart rate) automatically, but adaptation requires understanding individual biomechanics, motivation, injury risk, and real-time form correction—nuanced judgments that current AI systems cannot reliably make in a live group setting without human oversight. |
| Task automatability | claude-sonnet-5 | 2/5 | AI wearables and apps can track quantitative metrics like heart rate or reps, but adapting a program requires observing form, motivation, and physical cues in real time that current systems cannot fully replicate for in-person training. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Fitness instruction carries liability for exercise-related injury; trainers are often certified (ACE, NASM, etc.) and carry professional insurance; participants expect human judgment and personal attention, creating both regulatory and cultural barriers to full automation. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement generally applies to fitness training, but customer preference for human motivation/feedback and liability concerns around injury create moderate friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | A wearable-plus-analytics stack costs $50–500 annually per participant, but labor for a qualified trainer ($25–50/hour) remains far cheaper per participant in group settings when amortized across class size. |
| Cost vs. human wage | claude-sonnet-5 | 3/5 | App-based tracking and program adjustment is cheap per user, but it doesn't fully replace the human labor of live monitoring and coaching, so cost comparison is only partial when scoped to the full task. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Fitness tracking apps and wearables collect progress data, but no deployed product reliably adapts personalized programs in real time based on live observation; most systems require manual human review and decision-making to adjust programs safely. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Fitness apps (e.g., Whoop, Fitbod) offer basic progress tracking and auto-adjusted plans, but these are narrow, data-driven adjustments rather than the holistic in-session monitoring and adaptation a live instructor performs. |
Plan routines, choose appropriate music, and choose different movements for each set of muscles, depending on participants' capabilities and limitations.
32CI 25–39 · exposure 25 · augmentation 75 · importance 4.8/5 · click for rater detail
Plan routines, choose appropriate music, and choose different movements for each set of muscles, depending on participants' capabilities and limitations.
32| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Fitness instruction remains predominantly human-led with limited automation; while boutique studios and gyms use digital tools for scheduling and some form libraries, live instruction by humans is still the market norm, and adoption of autonomous routine planning is minimal. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Fitness services are a low-digitization, in-person sector where AI planning tools are used as aids by a minority of instructors; broad production-scale adoption is limited. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI can meaningfully assist instructors by generating routine drafts, suggesting music based on tempo and energy level, and offering exercise alternatives for different muscle groups—capabilities that can measurably raise instructor productivity while the human retains safety oversight and personalization authority. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI tools can meaningfully speed up brainstorming of routines, generate music selections, and suggest muscle-group variations, giving trainers a productivity boost while they retain final judgment. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | AI can assist in generating routine templates and music selection, but designing individualized routines that dynamically adapt to live participant capabilities, injuries, and real-time feedback requires constant human judgment and physical presence that current systems cannot replace end-to-end. |
| Task automatability | claude-sonnet-5 | 2/5 | AI can suggest exercise sequences and music playlists via generative tools, but tailoring to individual participants' physical capabilities, injuries, and real-time class dynamics requires embodied judgment AI lacks today.dominant |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Fitness instruction carries liability and safety concerns; participants expect and often legally require a credentialed human instructor present to assess limitations, modify movements in real-time, and supervise form—automation does not reduce these legal and organizational requirements. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement mandates a human specifically design routines, but liability concerns around injury from poorly matched movements and customer preference for human expertise create some friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | AI-powered playlist and routine-suggestion tools are relatively inexpensive, but the labor cost savings are minimal since human instructors must substantially customize and supervise output; the cost of oversight may approach the marginal instructor wage. |
| Cost vs. human wage | claude-sonnet-5 | 3/5 | AI-assisted planning tools are cheap and fast, but a human instructor still must personalize and adapt on the fly, so total cost savings versus a trainer doing this task themselves are moderate, not dramatic. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | While AI tools exist to suggest exercises and generate playlists, no deployed product reliably handles the full task of real-time personalization for diverse participant limitations in a live group setting with the nuance required for safety and effectiveness. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Some fitness apps and AI-generated workout planners exist, but they are narrow, generic, and not widely deployed for live group class customization with musical/movement pairing at scale. |
Evaluate individuals' abilities, needs, and physical conditions, and develop suitable training programs to meet any special requirements.
31CI 25–36 · exposure 25 · augmentation 75 · importance 4.5/5 · click for rater detail
Evaluate individuals' abilities, needs, and physical conditions, and develop suitable training programs to meet any special requirements.
31| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 3/5 | Fitness tech adoption is moderate: many gyms and studios have incorporated AI-assisted program design tools, but human instructors conducting assessments and developing individualized programs remain standard practice; automation is more common in consumer app space than in professional studio settings. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Fitness services sector has low digitization for hands-on physical assessment; AI adoption is mostly limited to app-based workout generators rather than embedded professional practice. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI tools substantially assist trainers by generating initial program templates, analyzing fitness data, and suggesting modifications based on user input, enabling faster program customization; however, the trainer must still validate recommendations and conduct the critical human assessment of physical condition and special needs. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI can meaningfully assist trainers by generating draft programs, tracking progress data, and suggesting modifications based on reported conditions, while the trainer retains responsibility for physical evaluation and judgment. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI can assist in analyzing fitness questionnaires and generating generic training program templates, the task requires nuanced evaluation of individual physical conditions, injury history, and adaptive needs that typically demand in-person physical assessment and real-time judgment that current AI systems cannot perform end-to-end. |
| Task automatability | claude-sonnet-5 | 2/5 | Physical assessment of abilities and conditions requires hands-on observation, movement screening, and interpersonal rapport that current AI cannot perform directly; only the program-writing portion is automatable, limiting overall time savings below the 50% threshold for the full task. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Liability and scope barriers are significant: fitness professionals face legal accountability if their program recommendations cause injury, and many jurisdictions require qualified human professionals to conduct fitness assessments and provide medical accommodations; organizational culture also favors in-person evaluation for program development. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No strict licensing mandate in most jurisdictions, but liability concerns around injury from poorly evaluated conditions and client expectation of hands-on assessment create meaningful friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | AI-generated fitness programs are cheap per unit, but the cost advantage erodes when accounting for the need for human review, liability oversight, and the inability of current systems to handle complex medical/physical assessments without human verification. |
| Cost vs. human wage | claude-sonnet-5 | 3/5 | AI-generated program templates are very cheap to produce, but the human assessment component still requires trainer time, keeping the blended cost closer to comparable rather than order-of-magnitude cheaper. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Some AI fitness apps can generate personalized workout plans from user input, but reliable systems for comprehensively evaluating special physical conditions (mobility limitations, past injuries, chronic conditions) and developing truly individualized programs remain limited; deployed products typically handle standard cases only. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Apps like fitness AI planners generate generic workout plans from questionnaire inputs, but reliable physical evaluation (posture, mobility, injury risk) in production remains rare and unvalidated at scale. |
Instruct participants in maintaining exertion levels to maximize benefits from exercise routines.
26CI 21–30 · exposure 17 · augmentation 75 · importance 4.5/5 · click for rater detail
Instruct participants in maintaining exertion levels to maximize benefits from exercise routines.
26| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Fitness centers and training studios are adopting wearables and app-based coaching, but primary delivery remains human-led classes; adoption is slow and augmentative rather than replacement-focused in mainstream fitness settings. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Fitness industry adoption of AI is nascent, mostly limited to app-based personal training and wearables rather than deep integration into group fitness instruction. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI wearables and fitness apps provide real-time feedback on heart rate, power output, and form cues that substantially assist instructors and participants in monitoring and maintaining exertion, raising engagement and results while the human instructor remains central to the experience. |
| Augmentation potential | claude-sonnet-5 | 4/5 | Wearables and apps providing real-time heart-rate and exertion feedback meaningfully help instructors and participants monitor and adjust effort levels during workouts. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI could provide generic cues about exertion levels via text or video, real-time monitoring and personalized instruction requires observation of individual form, effort, and physiological responses—demanding live human judgment and adaptation that current systems cannot reliably deliver end-to-end with 50% time saving. |
| Task automatability | claude-sonnet-5 | 1/5 | This requires real-time physical observation of a person's exertion, form, breathing, and biofeedback in a live setting, which current AI cannot perform end-to-end without physical presence or sensors integrated into a full system. |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | No explicit legal requirement for human instructors, but organizational preference for in-person instruction, member expectations, and the safety liability of automated exertion guidance create moderate friction against full substitution. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No licensing requirement mandates a human specifically for this task, but liability concerns around injury from improper exertion guidance and strong customer preference for in-person motivation create moderate friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Building a system to match a human trainer's ability to cue and adjust exertion in real time across diverse participants would require significant integration and oversight costs, likely exceeding the wage of a group fitness instructor per session taught. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | While app-based coaching is cheap, achieving the same quality of live, adaptive exertion monitoring and correction would require expensive sensor integration and human oversight, keeping costs comparable to a class instructor. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Some fitness apps and wearables offer exertion guidance, but deployed systems lack the nuance to truly instruct participants on maintaining optimal effort in real time; they rely on simplified metrics and do not reliably substitute for a human instructor's adaptive feedback. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Wearable-based apps and AI fitness coaches can suggest heart-rate zones, but no deployed product reliably instructs and adjusts exertion levels in real-time group settings at scale replacing human instructors. |
Teach proper breathing techniques used during physical exertion.
21CI 7–35 · exposure 13 · augmentation 50 · importance 4.2/5 · click for rater detail
Teach proper breathing techniques used during physical exertion.
21| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | While fitness tech adoption is growing, breathing instruction remains a core live-interaction element; adoption of AI replacements is minimal, with technology serving primarily as supplementary content rather than primary instruction. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Fitness instruction is a relatively low-digitization, in-person service sector where AI adoption for live coaching remains in early pilot stages (e.g., wearables, apps) rather than deep production use. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can assist instructors by generating instructional videos, curating breathing cue libraries, or providing pre-class planning materials, modestly raising instructor productivity without displacing the human's central teaching role. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI apps and wearables can supplement instruction with reminders, biofeedback, and generic breathing pattern guidance, but the instructor still primarily delivers hands-on, real-time coaching. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | Teaching breathing techniques requires real-time observation of individual form, immediate corrective feedback, and adaptation to each person's physical state and capability—tasks that demand live human presence and embodied responsiveness that current AI cannot replicate end-to-end. |
| Task automatability | claude-sonnet-5 | 2/5 | This requires real-time observation of a client's breathing pattern, physical form, and exertion level with immediate corrective cueing, which current AI cannot reliably perceive and correct in a live physical setting.“}, |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Physical fitness instruction typically requires human presence and direct observation; liability concerns around incorrect breathing instruction causing injury create strong organizational and legal friction against full automation. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement specifically bars AI from teaching breathing technique, but physical proximity, trust, and injury-risk oversight during exertion create moderate friction against pure automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | The integrated cost of AI systems (video generation, real-time monitoring, feedback loops, oversight) to teach breathing technique substitution would exceed the cost of a human instructor providing this service in live settings. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | While a video or app could deliver generic breathing instructions cheaply, achieving equal quality personalized coaching requires human presence and correction, keeping effective all-in AI cost not clearly cheaper. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | While AI can generate instructional videos or written guides on breathing techniques, no deployed product reliably teaches and corrects breathing form in real-time during live exercise classes or one-on-one training at production scale. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product coaches breathing technique during live physical exertion with hands-on or real-time visual correction; fitness apps offer only generic scripted guidance. |
Teach individual and team sports to participants through instruction and demonstration, using knowledge of sports techniques and of participants' physical capabilities.
21CI 11–30 · exposure 13 · augmentation 63 · importance 3.2/5 · click for rater detail
Teach individual and team sports to participants through instruction and demonstration, using knowledge of sports techniques and of participants' physical capabilities.
21| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | While fitness apps and digital instruction exist widely, adoption of AI-led or AI-replaced group instruction remains limited. Most commercial gyms and personal training still rely on human instructors; digital tools are adjunct, not replacement. Sectors are slow to substitute for the human coach role. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Fitness and personal training is a highly physical, in-person, moderately digitized sector with slow, uneven AI integration mostly limited to tracking and scheduling tools. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI can meaningfully assist trainers by providing instant form feedback via video analysis, generating customized programming, tracking participant metrics, and offering real-time prompts for cue delivery. These augmentations can measurably raise a human instructor's efficiency and personalization without removing them from the loop. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI can help trainers plan drills, analyze video of technique, and generate personalized programs, meaningfully aiding preparation even though delivery remains human-led. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI can generate exercise instructions, form correction via video analysis, and personalized workout plans at scale, the task fundamentally requires real-time demonstration, adaptive feedback based on individual observation, and motivational presence that current systems cannot reliably replicate end-to-end. Partial automation of program design or form assessment is feasible, but not the full teaching and demonstration function. |
| Task automatability | claude-sonnet-5 | 1/5 | This requires live physical demonstration, hands-on spotting, real-time adjustment to a person's body mechanics, and motivational presence that current AI cannot replicate in person. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Fitness instruction for health and safety has material liability exposure; errors in form or intensity can cause injury. Many participants prefer human interaction for motivation and real-time form correction. Organizational and legal friction around AI-driven coaching (especially for group settings) remains significant. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No licensing requirement typically, but liability for injury, need for physical presence, and client preference for human coaching create real friction against automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | High-quality personalized coaching AI (with video analysis and feedback) remains expensive per participant when accounting for infrastructure, integration, and liability oversight. It can undercut in-person instruction at scale only for pre-recorded or low-touch scenarios, not for the adaptive, safety-critical instruction described. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | While AI content generation is cheap, it cannot substitute for the actual coaching service, so effective cost comparison favors the human who delivers the full task. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Deployed products exist for video-based form feedback and workout app instruction, but they show material limitations in real-time adaptation, safety assessment for diverse participants, and the motivational/coaching components essential to group fitness. No mature system reliably replaces a live instructor's observational and corrective role. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product physically teaches sports techniques with demonstration and tactile correction; AI video apps offer static tips but not real embodied instruction. |
Organize and conduct competitions and tournaments.
19CI 9–30 · exposure 13 · augmentation 50 · importance 2.4/5 · click for rater detail
Organize and conduct competitions and tournaments.
19| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Fitness and group fitness sectors have adopted digital tools for scheduling and registration, but actual competition conduct remains human-led; adoption of autonomous competition management is minimal and shows no rapid displacement trend. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Fitness services are a low-digitization, high-physical-presence sector with minimal AI agent deployment for running live events. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can meaningfully assist with competition logistics (automated scoring, real-time leaderboards, participant tracking), but the human instructor retains primary responsibility for conducting and managing the event experience. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI tools can help with scheduling, bracket generation, participant tracking, and promotional materials, moderately aiding the organizational side of this task. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | Limited automation potential exists for scheduling, bracket generation, and online registration, but the core task of conducting live competitions requires real-time human judgment, motivation, and physical presence that current AI cannot replace end-to-end. |
| Task automatability | claude-sonnet-5 | 1/5 | Organizing and conducting live competitions/tournaments requires physical presence, real-time coordination, and on-site judgment that current AI cannot perform end-to-end.dll |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Significant barriers exist: organizational liability for event management, participant safety and legal responsibility, professional certification often required for fitness instruction, and strong preference for human leadership and motivation in live competitions. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No licensing requirement specifically for running competitions, but physical presence, liability for participant safety, and customer/participant expectation of human oversight create real friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | AI tools for event management and registration are inexpensive, but the human cost of organizing and conducting a competition (expertise, liability, physical presence) far exceeds what automation currently saves overall. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | Some administrative sub-tasks (bracket software, registration systems) are cheap, but the bulk of the task—on-site running of events—still requires paid human labor, making overall AI substitution not cheaper. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | While AI can assist with logistics (scheduling software, leaderboard management), no deployed product reliably conducts competitions autonomously; human facilitators remain essential for rule enforcement, participant engagement, and real-time decision-making. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed AI product runs physical fitness competitions or tournaments autonomously; at most software assists with scheduling or bracket generation. |
Explain and enforce safety rules and regulations governing sports, recreational activities, and the use of exercise equipment.
13CI 9–16 · exposure 5 · augmentation 38 · importance 4.4/5 · click for rater detail
Explain and enforce safety rules and regulations governing sports, recreational activities, and the use of exercise equipment.
13| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Fitness and recreation remain highly human-centric, hands-on industries. Although some facilities use digital monitoring aids, adoption of AI-driven safety enforcement is minimal and slow, with organizational preference for in-person instructors bearing explicit accountability. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Fitness and recreational services are a low-digitization, physically-oriented sector with minimal AI agent deployment for real-time safety enforcement. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can assist by flagging poor form via pose detection, suggesting safety cues, or reminding participants of rules via recorded announcements, materially helping instructors manage multiple participants. However, the human instructor must remain the primary enforcer and decision-maker. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI could help draft safety policies, generate training materials, or provide checklists, but offers little real-time assistance during actual safety enforcement and monitoring. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | This task requires real-time monitoring of human behavior, physical form correction, and dynamic judgment about safety hazards in a live environment. Current AI cannot reliably perform end-to-end enforcement of safety in physical spaces with the situational awareness and immediate intervention capacity needed. |
| Task automatability | claude-sonnet-5 | 1/5 | This requires real-time physical presence to observe clients' form, spot dangerous equipment use, and intervene immediately, which current AI cannot perform end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Fitness facilities face significant liability exposure if they delegate safety enforcement to unmonitored AI systems. Regulatory oversight, duty-of-care standards, and insurance requirements create legal barriers to full automation; a human instructor must remain responsible for participant safety. |
| Adoption barriers | claude-sonnet-5 | 4/5 | Liability concerns around injury, gym insurance requirements, and the need for a physically present person to intervene in real-time create strong practical and legal barriers to full automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Developing and maintaining AI monitoring systems for a gym or fitness facility (hardware, software, integration, human oversight) would likely exceed the cost of direct instructor supervision, especially given liability exposure and the need for human judgment. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | Since AI cannot substitute for the physical enforcement component, any AI solution would only supplement rather than replace the human, making cost comparison largely inapplicable or requiring continued human wages plus AI tool costs. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | While AI can generate written safety guidelines and static educational content, no deployed product reliably monitors and enforces safety rules in live fitness settings. Some computer vision systems exist for pose correction, but they lack the contextual judgment and liability readiness for safety enforcement. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product physically monitors and enforces safety rules in a gym or recreational setting; this remains a research/theoretical possibility only, not a production capability. |
Conduct therapeutic, recreational, or athletic activities.
12CI 7–16 · exposure 8 · augmentation 50 · importance 3.8/5 · click for rater detail
Conduct therapeutic, recreational, or athletic activities.
12| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Adoption of AI in fitness remains slow and limited to supplementary tools (app-based coaching, form analysis). In-person group fitness instruction remains a predominantly human-led activity; gyms and clinics have not systematically replaced instructors with AI agents. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Fitness and recreational services are a physically-delivered, lower-digitization sector where AI adoption for actual activity conduct remains minimal, though apps for planning are growing. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can assist human instructors by analyzing participant form via computer vision, generating personalized workout plans, or tracking progress metrics, improving the instructor's ability to tailor sessions. However, the primary task of conducting and leading activities remains human-centered. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI can help design workout plans, track progress, and suggest activity variations, offering moderate assistance to trainers even though it cannot conduct the sessions itself. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | AI cannot physically conduct activities or provide real-time in-person instruction. While AI could design workout programs or provide remote guidance via video, the core task of actively leading, demonstrating, and adjusting for individual participants in real time requires human presence and physical demonstration. |
| Task automatability | claude-sonnet-5 | 1/5 | This task requires live physical presence, demonstration, hands-on spotting/correction, and real-time motivational engagement that current AI cannot perform end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Therapeutic and athletic activities, especially in clinical or professional settings, often require licensed practitioners and direct human-client contact. Liability, safety, injury risk, and regulatory requirements for exercise instruction (particularly in therapeutic contexts) create significant legal and organizational barriers to full automation. |
| Adoption barriers | claude-sonnet-5 | 4/5 | Physical safety, liability for injury, and often certification/licensing requirements for instructing exercise or therapeutic activities create strong barriers to full substitution. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | AI systems cannot deliver the in-person, real-time leadership of group fitness at lower cost than a human instructor. The human instructor's wage remains the baseline; AI tools may assist but do not eliminate the need for the instructor's presence. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | AI cannot substitute for the physical delivery of the activity, so there is no viable AI cost comparison for the core task itself. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed product reliably performs the actual conduct of therapeutic, recreational, or athletic activities with humans. While fitness apps and AI coaching exist, they do not replace the live instructor's role of leading group sessions, monitoring form, and providing immediate corrections. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product actually conducts in-person therapeutic, recreational, or athletic activities; existing apps only provide instructional content, not the physical conduct of the session. |
Advise participants in use of heat or ultraviolet treatments and hot baths.
11CI 0–23 · exposure 8 · augmentation 25 · importance 2.2/5 · click for rater detail
Advise participants in use of heat or ultraviolet treatments and hot baths.
11| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Fitness instruction remains a highly human-contact, in-person sector with low automation adoption; participants expect and rely on immediate, embodied feedback from live instructors, and regulatory/liability concerns slow any shift toward autonomous AI advisors. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Fitness and wellness services are a low-digitization, physically-delivered sector with minimal AI agent deployment for hands-on safety advising; adoption in this specific niche is essentially nonexistent. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | AI could generate templated safety talking points or contraindication checklists for an instructor to review, but the core task—observing individuals, assessing fit, and advising in real time—relies on human judgment and presence that AI augmentation offers only marginal support for. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI chatbots could supplement a trainer's knowledge base with general safety guidelines on heat/UV exposure, offering modest informational support, but this is a narrow slice of the full advising task. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | This task requires real-time human interaction, assessment of individual participant conditions, contraindications, and personalized safety advice that depends on observing physical responses and building rapport—elements current AI cannot reliably perform in a live group setting. |
| Task automatability | claude-sonnet-5 | 2/5 | AI can provide generic informational guidance on heat/UV/hot bath use, but real advising requires observing the participant, adjusting for physical condition, and hands-on supervision, which AI cannot perform end-to-end today.4 Text-based advice alone doesn't meet the full task scope of in-person guidance. |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Health and safety advice, especially regarding heat and UV exposure, falls under implicit medical/therapeutic oversight; instructors carry professional liability, and most fitness facilities require a human instructor to be present and responsible for participant safety guidance. |
| Adoption barriers | claude-sonnet-5 | 3/5 | There's no licensing requirement specific to this narrow task, but liability concerns (burns, heat stroke, skin damage from UV) and facility safety policies create meaningful friction against pure AI substitution without human oversight. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | AI inference for advisory text is cheap, but the real cost driver is integration into group fitness operations, liability management, and oversight—which together exceed the hourly wage of a fitness instructor for this task. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | AI could cheaply generate generic safety information, but since the task requires physical presence and judgment about a specific person's condition, a human trainer's cost isn't meaningfully displaced, keeping the effective cost ratio unfavorable to AI. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed AI product performs live, in-person health and safety advice for heat/UV treatments in a fitness class context; any such system would face regulatory barriers and would lack the embodied presence and liability coverage a human instructor provides. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed fitness product provides physical supervision of participants using heat, UV, or hot bath treatments; this remains an in-person, physically supervised activity not addressed by production AI systems. |
Maintain fitness equipment.
10CI 5–15 · exposure 0 · augmentation 25 · importance 4.1/5 · click for rater detail
Maintain fitness equipment.
10| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Fitness facilities are not adopting AI for equipment maintenance because the task is inherently physical; adoption of any kind remains negligible in this domain. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Fitness and physical facility maintenance is a low-digitization, physical-labor sector with minimal AI adoption for hands-on equipment upkeep. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | AI could assist minimally by logging maintenance schedules or suggesting when maintenance is needed based on usage data, but such assistance is limited compared to the hands-on expertise required for actual repair work. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI could assist with scheduling maintenance reminders or diagnosing equipment issues via sensor data, but this offers only limited assistance to the core physical maintenance task. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | Maintaining fitness equipment requires physical inspection, hands-on repair, mechanical troubleshooting, and judgment about wear patterns—tasks that current AI cannot perform end-to-end without direct physical manipulation capabilities, which deployed systems lack. |
| Task automatability | claude-sonnet-5 | 1/5 | Maintaining fitness equipment involves physical inspection, cleaning, tightening, lubricating, and repairing machines, which requires manual dexterity and physical presence that current AI cannot replicate.' |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Equipment maintenance may require certification or warranty-authorized technician status in many gym facilities, and liability concerns around faulty repairs create organizational and legal barriers to substitution. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement exists for this task, but the physical nature of maintenance (using tools, physical inspection) creates a practical barrier to AI substitution rather than a regulatory one. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | AI systems have no meaningful cost advantage here since the task cannot be automated; a human technician must physically inspect and repair equipment, making the loaded human wage the baseline. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no AI system substitute for this physical task, so any AI-based cost comparison is not applicable; human labor remains the only viable and cost-effective option. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed AI system can reliably perform equipment maintenance autonomously; the task fundamentally requires physical action and real-time diagnosis on complex machinery that current AI agents cannot access or manipulate. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed AI product performs physical maintenance of gym equipment; this remains a purely manual, hands-on task performed by humans or maintenance technicians. |
Offer alternatives during classes to accommodate different levels of fitness.
9CI 5–13 · exposure 0 · augmentation 25 · importance 4.8/5 · click for rater detail
Offer alternatives during classes to accommodate different levels of fitness.
9| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Fitness instruction remains a sector with strong demand for in-person, live human instructors who build community and provide real-time motivation. While some studios use pre-recorded videos, autonomous or agent-based real-time fitness class adaptation is not in measurable production adoption. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Fitness instruction is a physical, low-digitization service sector where AI adoption for live class delivery remains minimal despite some app-based fitness tools. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | AI could assist instructors by pre-planning exercise variations or suggesting alternatives during preparation, but current systems offer limited real-time support during live class delivery. The augmentation is minor because the core task (dynamic, in-the-moment accommodation) remains almost entirely human-dependent. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI can help instructors plan workout variations or modifications in advance, but offers little real-time assistance during the live class itself. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | This task requires real-time observation of individual participants, assessment of their physical condition, and adaptive verbal/visual communication during a live class. Current AI systems cannot reliably perceive the fitness level of multiple people simultaneously in a dynamic environment and provide personalized alternatives while maintaining class flow and motivation. |
| Task automatability | claude-sonnet-5 | 1/5 | This requires real-time physical observation of participants' form, fatigue, and capability during live movement, which current AI cannot perceive and adapt to in a physical group class setting. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Fitness instruction involves direct physical presence, safety liability for exercise form corrections, and participant trust/motivation that strongly prefer human instructors. Liability for injury if AI-suggested modifications are inappropriate and the expectation of human engagement create significant adoption barriers. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No licensing requirement mandates a human specifically, but the need for physical presence, liability for injury, and real-time judgment create substantial practical barriers to automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | An AI system capable of real-time participant monitoring, exercise adaptation, and live instruction would require expensive computer vision infrastructure, motion analysis models, and interactive AI agents. This cost far exceeds the wage of a group fitness instructor offering modifications. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no viable AI substitute performing this in-person adaptive coaching function, so cost comparison favors the human instructor entirely. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed product today autonomously leads a fitness class and adapts exercise variations in real time based on live observation of participant capabilities. This requires embodied presence, real-time perception, and social-emotional interaction that current AI systems cannot execute reliably in production. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product offers real-time in-person modification cues for group fitness classes based on live observation of participant ability. |
Teach and demonstrate use of gymnastic and training equipment, such as trampolines and weights.
9CI 5–13 · exposure 5 · augmentation 38 · importance 4.5/5 · click for rater detail
Teach and demonstrate use of gymnastic and training equipment, such as trampolines and weights.
9| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Fitness and training sectors show slow adoption of full automation; they remain highly reliant on in-person instructors for member engagement, retention, and liability management, though video and hybrid offerings are growing from a low base. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Fitness instruction is a low-digitization, physically-embodied service sector with minimal AI/robotic adoption for hands-on equipment instruction. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can assist by generating instructional video libraries, providing form-correction suggestions via computer vision, or creating personalized workout plans that trainers then refine and supervise in person, improving some instructor workflows without removing the human. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI video analysis apps or wearables can offer some form-feedback, but they provide limited assistance compared to a live instructor demonstrating equipment use. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | This task requires real-time physical demonstration, direct safety oversight, and immediate correction of form to prevent injury. Current AI cannot physically demonstrate exercises or provide real-time in-person safety monitoring, which are core to the task. |
| Task automatability | claude-sonnet-5 | 1/5 | This requires live physical demonstration, hands-on spotting, and real-time correction of body mechanics on physical equipment, which current AI cannot perform end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Strong barriers exist: liability exposure is high if AI-generated form correction leads to injury, fitness facilities face legal responsibility for participant safety, and there is inherent expectation of human-led supervision in group fitness contexts that regulatory bodies and insurance carriers reinforce. |
| Adoption barriers | claude-sonnet-5 | 4/5 | Safety liability (trampolines, weights) creates strong practical requirements for a present, qualified human instructor to prevent injury, even without formal licensing mandates in most jurisdictions. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | AI video generation and instructional tools have setup costs and require human review for safety accuracy. The total cost per instance, including quality oversight, exceeds what a human instructor would charge for in-person or pre-recorded group instruction. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | AI has no physical embodiment to perform this task, so there is no viable cost comparison—human instruction remains the only functional option. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | While AI can create video instructional content or generate exercise descriptions, deployed systems lack the ability to physically demonstrate equipment use or supervise participants in real time, which are essential elements of teaching and demonstrating. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product physically demonstrates equipment use or spots trainees; this remains firmly outside current AI product capability. |
Administer emergency first aid, wrap injuries, treat minor chronic disabilities, or refer injured persons to physicians.
3CI 0–5 · exposure 5 · augmentation 25 · importance 3.7/5 · click for rater detail
Administer emergency first aid, wrap injuries, treat minor chronic disabilities, or refer injured persons to physicians.
3| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Fitness and wellness sectors rely on in-person, certified instructors and are slow to digitize core health and safety functions; regulatory and liability concerns prevent automation of emergency medical tasks. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Fitness and physical training sectors show minimal AI adoption for hands-on physical care tasks; this is a low-digitization, physically embodied task category. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | AI could marginally assist by providing quick reference information (injury classification, referral pathways) or video analysis of mechanics, but the human instructor remains the decision-maker and primary agent throughout. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI could provide reference information (e.g., first-aid protocols via an app) or triage guidance, but offers minimal assistance for the actual physical administration of care. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | This task requires direct physical contact, real-time assessment of injury severity, hands-on intervention (wrapping, bandaging), and medical judgment about referral appropriateness—all demanding human presence and tactile engagement that current AI cannot perform end-to-end. |
| Task automatability | claude-sonnet-5 | 1/5 | This requires hands-on physical intervention (wrapping injuries, administering first aid) that current AI systems cannot perform, lacking any physical embodiment for real-world manipulation.-based tasks. |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Strong legal and safety barriers exist: a trained, often certified human (CPR/first aid certification) must be present and perform or directly supervise emergency care; liability and duty-of-care standards require human accountability and hands-on intervention. |
| Adoption barriers | claude-sonnet-5 | 5/5 | Emergency first aid and injury treatment involve liability, certification requirements (e.g., CPR/first aid certification), and immediate physical safety concerns that legally and practically require a present, qualified human. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | The labor cost for trained fitness instructors to deliver emergency care and injury management is already low; AI integration would require expensive human oversight and liability coverage without replacing the need for onsite human presence. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | AI cannot perform the physical components at all, so there is no viable cost comparison — a human must be present and physically capable. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | AI systems can provide decision-support (symptom checkers, referral guidance) or video analysis of injury mechanisms, but no deployed product reliably performs the core physical first aid, injury wrapping, or differential assessment needed in this task at production scale. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed AI product administers physical first aid or wraps injuries; this remains purely a human physical-manual task with no robotic deployment in this context. |
Related occupations — Personal Care & Service
How to read this
A high substitution score does not mean this job disappears — it means a large share of its current tasks face replacement pressure, so the mix of tasks is likely to change. High augmentation alongside substitution typically means the occupation reorganizes around the protected tasks. Wide confidence intervals mean the rater panel disagreed: treat those scores as open questions, not verdicts.
What would change this score
New model capabilities (automatability, feasibility), falling inference costs (cost ratio), regulation and licensing shifts (barriers), and measured sector adoption (velocity) all re-enter at every index release. Each release is recomputed, versioned and kept queryable — scores are claims with a date on them, not permanent labels.