Lifeguards, Ski Patrol, and Other Recreational Protective Service Workers

33-9092.00
Median wage $33,580/yr157,550 employed (US)Rank #834 of 923 scored · top 90% by substitution

Monitor recreational areas, such as pools, beaches, or ski slopes, to provide assistance and protection to participants.

Sub-scores

0–100 · band = confidence interval from rater disagreement

Substitution13
Exposure11
Augmentation34

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

15 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%12

panel mean rating 1.5/5 → substitution pressure 12/100

Technical feasibility todayw 20%11

panel mean rating 1.4/5 → substitution pressure 11/100

Cost vs. human wagew 15%13

panel mean rating 1.5/5 → substitution pressure 13/100

Adoption barriersw 20%inverted — strong barriers lower the score22

panel mean rating 4.1/5 (barrier strength) → substitution pressure 22/100

Sector adoption velocityw 10%7

panel mean rating 1.3/5 → substitution pressure 7/100

Task breakdown (15 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.

Maintain quality of pool water by testing chemical levels.

51

CI 30–72 · exposure 50 · augmentation 63 · importance 4.3/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510013/5Automated pool monitoring is common in larger public facilities and commercial operations, but adoption remains uneven—many smaller pools and recreational centers still rely on manual testing, reflecting the mixed digitization profile of the leisure and hospitality sector.
Sector adoption velocityclaude-sonnet-52/5Recreational and facilities-management sectors are slow adopters of AI/automation generally, with pool chemical monitoring automation existing mainly in high-end or commercial aquatic facilities rather than widespread deployment.
Augmentation potentialclaude-haiku-4-5-202510014/5AI-driven monitoring systems significantly assist lifeguards by providing continuous real-time alerts and trend data, allowing staff to focus on safety supervision rather than routine testing; this augmentation improves response time to chemical imbalances while keeping the human in oversight.
Augmentation potentialclaude-sonnet-53/5Digital water-testing devices and connected sensors can assist lifeguards by providing faster, more accurate readings and alerts, improving efficiency of the monitoring task while the human remains responsible for verification and corrective action.
Task automatabilityclaude-haiku-4-5-202510014/5Testing chemical levels is a highly structured task with measurable inputs (pH, chlorine, alkalinity) and clear thresholds; current automated pool monitoring systems can continuously measure and log these metrics, achieving significant time savings compared to manual testing. However, interpretation of anomalies and corrective actions may still require human judgment, preventing a full 5 rating.
Task automatabilityclaude-sonnet-52/5Automated pool chemistry sensors and dosing systems exist and can measure/adjust chemical levels, but the task as performed by this worker typically includes physical sampling, verification, and judgment calls tied to a physical space that AI alone cannot fully replace end-to-end today.
Adoption barriersclaude-haiku-4-5-202510013/5While there are no strict licensing barriers preventing automation of the measurement itself, health and safety regulations often require documented human oversight and sign-off on water quality; some facilities may also prefer human surveillance of the pool environment simultaneously, introducing organizational friction.
Adoption barriersclaude-sonnet-52/5Health and safety codes often require certified staff to verify water safety and maintain logs, creating some regulatory/liability friction, though this is not a hard licensing barrier preventing automated sensors from assisting.
Cost vs. human wageclaude-haiku-4-5-202510014/5Automated water testing systems (sensors, logging, alerts) have declining hardware and maintenance costs that are substantially lower than the hourly wage of a lifeguard performing frequent manual testing rounds; the cost advantage grows with pool size and testing frequency.
Cost vs. human wageclaude-sonnet-52/5Automated dosing/sensor systems require significant upfront capital investment, installation, and maintenance, so for many facilities the human performing manual tests remains cheaper or comparable, especially at small-scale operations.
Technical feasibility todayclaude-haiku-4-5-202510014/5Automated pool monitoring systems are commercially deployed and used in production at many facilities; they reliably measure chemical parameters. The task is mature in technical capability, though integration and oversight still typically involve human staff confirmation rather than full autonomy.
Technical feasibility todayclaude-sonnet-52/5Automated chemical monitoring/dosing hardware is commercially deployed in some facilities, but it is a hardware/IoT solution rather than a general AI system, and many pools still rely on manual test-strip or reagent testing performed by staff.

Complete and maintain records of weather and beach conditions, emergency medical treatments performed, and other relevant incident information.

32

CI 23–43 · exposure 33 · augmentation 50 · importance 4.3/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Recreational protective services remain largely traditional and low-digitization sectors with small, decentralized organizations. While some large resorts and municipalities pilot digital record systems, adoption is slow and hesitant, with many still relying on paper logs or basic spreadsheets.
Sector adoption velocityclaude-sonnet-51/5Recreational safety services (beaches, ski patrols) are a low-digitization, physically dispersed, small-organization sector with minimal AI tooling adoption reported to date.
Augmentation potentialclaude-haiku-4-5-202510013/5AI could assist by suggesting structured fields, flagging unusual patterns in conditions or incident frequencies, and auto-populating routine weather data from external APIs. This would streamline the human's documentation process, though the person remains essential for judgment and legal accountability.
Augmentation potentialclaude-sonnet-53/5AI transcription and templated form-generation could meaningfully speed up documentation and reduce errors, though the underlying incident details must still be observed and reported by humans.
Task automatabilityclaude-haiku-4-5-202510012/5While AI could potentially structure and store recorded observations, the task requires real-time judgment about dangerous conditions, interpretation of equipment readings, and contextual medical assessment. Most of the interpretive work cannot be automated reliably, though data entry and formatting could see modest automation.
Task automatabilityclaude-sonnet-53/5Structured record-keeping of weather, beach conditions, and incident/medical details is largely a data-entry and summarization task well-suited to AI, though data collection (observing conditions, treating patients) requires physical human presence before recording.
Adoption barriersclaude-haiku-4-5-202510014/5Legal and liability barriers are high: incident records are often evidence in liability claims, worker compensation cases, and regulatory inspections. Regulatory bodies (OSHA, state water safety boards) typically require authorized personnel to certify accuracy; an AI-only system would face resistance and may not satisfy legal admissibility standards.
Adoption barriersclaude-sonnet-53/5Incident records, especially those involving medical treatment, often have legal/liability significance requiring accurate human-verified documentation, creating moderate barriers to full automation of record completion.
Cost vs. human wageclaude-haiku-4-5-202510012/5The cost of an AI system to capture, process, and validate this data (including integration with beach/ski patrol operations and error correction) likely exceeds the low cost of a lifeguard or patrol officer spending 15–30 minutes daily on documentation.
Cost vs. human wageclaude-sonnet-53/5Voice-to-text and templated reporting tools could cut documentation time cheaply, but integration with existing incident management systems and verification adds cost, making savings moderate rather than dramatic.
Technical feasibility todayclaude-haiku-4-5-202510012/5No deployed product reliably captures and organizes the complex, contextual information required (weather nuance, incident severity, medical details) without substantial human oversight and correction. Form-filling and template systems exist but don't handle the full scope without error-prone manual review.
Technical feasibility todayclaude-sonnet-52/5While generic dictation, form-filling, and note-summarization tools exist, no deployed product specifically integrates weather sensors, beach observations, and incident logging for lifeguard/ski patrol operations at scale today.

Provide assistance with staff selection, training, and supervision.

28

CI 25–30 · exposure 25 · augmentation 50 · importance 3.7/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Small and mid-sized recreational facilities and ski patrols have low digital sophistication and minimal adoption of AI HR tools; adoption remains in the pilot phase rather than production deployment across these labor-intensive, often seasonal operations.
Sector adoption velocityclaude-sonnet-52/5Recreational safety services (lifeguarding, ski patrol) are a low-digitization, physically-oriented sector with minimal reported AI adoption for HR-related functions.
Augmentation potentialclaude-haiku-4-5-202510013/5AI can usefully assist with resume screening, candidate pool analysis, and training content generation, helping HR staff work more efficiently, but the core supervisory and judgment functions remain human-driven and the augmentation is moderate rather than transformative.
Augmentation potentialclaude-sonnet-53/5AI can assist with generating training content, scheduling, and administrative aspects of staff management, improving efficiency while humans retain core supervisory responsibility.
Task automatabilityclaude-haiku-4-5-202510012/5AI can assist with drafting job descriptions, resume screening, and training material generation, but the critical judgment of assessing candidate fit, evaluating supervisor competence, and real-time supervision decisions require human expertise and accountability that current systems cannot reliably replace end-to-end.
Task automatabilityclaude-sonnet-52/5AI can help draft training materials, job postings, and schedules, but actual selection judgment, supervision, and interpersonal management require human decision-making and oversight that AI cannot fully replace end-to-end.'
Adoption barriersclaude-haiku-4-5-202510014/5Supervisory authority, duty of care liability, and accountability for staff competence in safety-critical recreational settings create strong legal and organizational barriers to full automation; human managers must retain final decision authority and sign-off on hiring and training decisions.
Adoption barriersclaude-sonnet-53/5No formal licensing requirement blocks AI use, but supervisory and personnel decisions typically require human accountability and carry legal/liability implications for staffing errors.
Cost vs. human wageclaude-haiku-4-5-202510012/5AI-assisted recruitment and learning platforms reduce some HR overhead, but the supervision and adaptive training components still require significant human involvement, keeping total costs in the same ballpark as traditional HR labor rather than an order of magnitude cheaper.
Cost vs. human wageclaude-sonnet-52/5Generic HR/AI tools have low marginal cost, but the human judgment and accountability components of supervision and training still require paid staff time, keeping overall costs comparable to human-led processes.
Technical feasibility todayclaude-haiku-4-5-202510012/5Deployed HR systems can automate resume parsing and training module delivery, but no mature product reliably handles the full pipeline of staff selection judgment, adaptive training customization, or performance supervision in recreational safety contexts where error costs are high.
Technical feasibility todayclaude-sonnet-52/5HR software and AI-assisted hiring tools exist and are used in some contexts, but for small recreational safety teams there's little evidence of deployed products handling this specific supervisory task reliably.

Inspect recreational facilities for cleanliness.

25

CI 23–28 · exposure 20 · augmentation 38 · importance 3.6/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Recreational facilities and ski patrol operations are typically small to medium organizations with lower digitization and slower AI adoption patterns. Adoption of automated inspection systems remains minimal in these sectors.
Sector adoption velocityclaude-sonnet-51/5Recreational and facility services are a low-digitization, physical-labor sector with minimal AI adoption for routine physical inspection tasks.
Augmentation potentialclaude-haiku-4-5-202510013/5AI-assisted visual inspection tools could help workers identify areas needing attention faster and more systematically, moderately improving inspection productivity while the human remains responsible for final assessment and corrective decisions.
Augmentation potentialclaude-sonnet-52/5Checklists, apps, or camera-based anomaable detection could assist by flagging areas needing attention, but such tools are not widely used for this specific task today.
Task automatabilityclaude-haiku-4-5-202510012/5Visual inspection for cleanliness can be partially automated using computer vision to detect obvious debris or contamination, but nuanced assessments of facility condition (e.g., subtle hygiene issues, wear patterns) and decision-making about safety standards require human judgment. Current AI cannot reliably meet the 50% time-saving bar for comprehensive facility inspection.
Task automatabilityclaude-sonnet-52/5Visual inspection of physical facilities requires embodied presence and judgment about cleanliness/safety hazards that current AI cannot perform end-to-end without robotics or extensive camera infrastructure.
Adoption barriersclaude-haiku-4-5-202510013/5Health and safety regulations often require human inspection and sign-off, and facility liability concerns create moderate friction against full automation. However, the task itself is not legally required to be performed by a licensed professional, allowing for augmentation.
Adoption barriersclaude-sonnet-53/5No formal licensing barrier for inspection itself, but liability for missed hazards (slips, contamination) and the need for a physically present responsible person creates moderate organizational friction against full automation.
Cost vs. human wageclaude-haiku-4-5-202510012/5Computer vision infrastructure, integration, ongoing maintenance, and human oversight for verification make the all-in cost comparable to or potentially higher than hiring workers for routine inspections, particularly in smaller facilities.
Cost vs. human wageclaude-sonnet-52/5Without robotic or sensor infrastructure, AI cannot substitute for a human walking the facility, so all-in costs of any AI-based system (cameras, sensors, monitoring software) likely exceed simply having staff inspect during their shift.
Technical feasibility todayclaude-haiku-4-5-202510012/5While computer vision systems exist for basic visual inspection, no mature deployed product reliably performs comprehensive cleanliness inspection in recreational facilities at production scale. Systems are research-stage or limited to narrowly scoped use cases like detecting trash in specific areas.
Technical feasibility todayclaude-sonnet-51/5No deployed product autonomously performs walk-through cleanliness inspections of recreational facilities like pools, decks, or ski areas; this remains a manual, physical task.

Inspect recreational equipment, such as rope tows, T-bars, J-bars, or chair lifts, for safety hazards and damage or wear.

14

CI 5–23 · exposure 13 · augmentation 38 · importance 4.0/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Ski resorts and recreational facilities operate in a conservative, regulated environment with strong institutional inertia. Adoption of AI for safety-critical inspections is minimal; organizations rely on established certification and human expertise, and regulatory frameworks do not yet permit autonomous machine-only sign-off.
Sector adoption velocityclaude-sonnet-51/5Recreational and outdoor safety sectors have low digitization and are laggards in AI adoption, especially for physical equipment inspection tasks.
Augmentation potentialclaude-haiku-4-5-202510013/5AI-assisted visual flagging (highlighting wear patterns or anomalies detected in pre-inspection imagery) could accelerate a human inspector's workflow and reduce oversight time, but the inspector remains essential for judgment, hands-on checks, and liability sign-off.
Augmentation potentialclaude-sonnet-52/5AI could potentially assist with logging inspection results, scheduling, or flagging patterns from sensor data, but offers minimal direct assistance to the physical inspection process itself.
Task automatabilityclaude-haiku-4-5-202510012/5AI vision systems could identify some visible damage or wear patterns in images, but inspecting complex mechanical systems like chair lifts requires hands-on assessment of structural integrity, wear mechanics, and safety tolerances that current AI cannot reliably perform end-to-end. The task involves physical interaction and contextual judgment beyond what visual inspection alone provides.
Task automatabilityclaude-sonnet-51/5This requires physical, hands-on inspection of mechanical equipment for wear and damage, involving tactile and visual assessment in real-world environments that current AI cannot perform end-to-end.
Adoption barriersclaude-haiku-4-5-202510014/5Strong regulatory and liability barriers exist: most jurisdictions require certified, trained human inspectors to sign off on lift safety before operation. Failure to detect hazards carries catastrophic liability, and liability law typically requires human professional judgment for equipment that affects public safety.
Adoption barriersclaude-sonnet-54/5Safety-critical mechanical equipment inspections are typically subject to regulatory and liability requirements mandating qualified human inspection and sign-off, creating strong barriers to automation.
Cost vs. human wageclaude-haiku-4-5-202510012/5AI vision inference is cheap, but integration, validation, liability insurance, and human oversight for safety-critical decisions add substantial cost. The loaded wage of a ski patrol inspector or lifeguard is moderate, and the total cost of a reliable AI system (including false-negative penalties) likely exceeds the human baseline.
Cost vs. human wageclaude-sonnet-51/5There is no viable AI substitute performing this physical inspection task, so cost comparison favors the human worker who can actually complete the task.
Technical feasibility todayclaude-haiku-4-5-202510012/5Computer vision can detect surface-level defects in photographs, but no deployed product reliably performs end-to-end safety inspections of operational lift systems. Existing AI vision systems have not been validated for liability-critical safety sign-off in recreational equipment contexts, and real-world deployment requires integration with maintenance workflows.
Technical feasibility todayclaude-sonnet-51/5No deployed AI product autonomously inspects ski lift mechanical components for safety hazards; this remains a physical inspection task requiring human presence and judgment.

Observe activities in assigned areas, using binoculars, to detect hazards, disturbances, or safety infractions.

13

CI 0–25 · exposure 13 · augmentation 38 · importance 4.2/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Adoption of AI-driven lifeguard assistance is slow. Most pools, beaches, and ski patrols still rely on human observation; pilot programs with AI monitoring are rare and typically augmentative rather than substitutive. Sectors remain conservative due to liability and regulatory constraints.
Sector adoption velocityclaude-sonnet-51/5Recreational safety and physical security sectors show minimal AI adoption for autonomous hazard monitoring; this remains a manual, low-digitization occupation.
Augmentation potentialclaude-haiku-4-5-202510013/5AI can usefully assist by flagging potential hazards (unusual clustering, sudden immobility in water) or helping scan large areas on camera feeds, prompting human verification and reducing observer fatigue. However, the core task of judgment and emergency response remains human-led, limiting the transformation.
Augmentation potentialclaude-sonnet-52/5Some camera-assisted drowning detection or motion-sensor alert tools can supplement human vigilance, but adoption is limited and the human remains the primary observer.
Task automatabilityclaude-haiku-4-5-202510012/5While AI-powered video monitoring can detect some hazards (e.g., people in water, crowd density anomalies), the task requires real-time discrimination between normal recreation and genuine emergencies, judgment of risk context, and identification of subtle safety infractions. Current systems struggle with the nuanced environmental understanding and false-positive rates that would meet the 50% time-saving bar for equal quality in a safety-critical context.
Task automatabilityclaude-sonnet-51/5This requires real-time physical presence, visual monitoring of dynamic outdoor/water environments, and immediate judgment-based intervention capability that current AI cannot replicate end-to-end without embodiment.
Adoption barriersclaude-haiku-4-5-202510014/5Strong barriers exist: liability risk is extreme if automated systems miss emergencies, most jurisdictions require a licensed/present lifeguard to observe and respond, organizational culture and insurance requirements mandate human oversight, and parent/visitor expectations create friction against full automation. Legal duty to provide trained human supervision is typically non-delegable.
Adoption barriersclaude-sonnet-55/5Lifeguarding and ski patrol are safety-critical, often require certification and legal duty-of-care standards, and liability for failure to prevent injury/death creates strong barriers to full automation.
Cost vs. human wageclaude-haiku-4-5-202510012/5A lifeguard's loaded wage is modest (~$30k–$40k annually), while a robust surveillance system with AI backbone, hardware, integration, and monitoring oversight can easily cost $50k–$150k+ to install and maintain, making the cost-per-observation unfavorable relative to human labor.
Cost vs. human wageclaude-sonnet-51/5Deploying cameras, sensors, and monitoring infrastructure plus required human backup for liability is more expensive than paying a single lifeguard, especially at small facilities.
Technical feasibility todayclaude-haiku-4-5-202510012/5Computer vision systems can perform object and activity detection in production (e.g., crowd monitoring), but deploying them reliably for lifeguard-level hazard detection—distinguishing genuine distress from playful behavior, detecting infractions across varied lighting and water conditions—remains materially error-prone. Few organizations deploy fully autonomous monitoring; most systems are still in pilot or supplementary roles.
Technical feasibility todayclaude-sonnet-51/5No deployed product autonomously performs lifeguard/ski patrol surveillance and hazard detection with human-equivalent reliability; camera-based drowning detection systems exist only as narrow pilot aids, not replacements.

Instruct participants in skiing, swimming, or other recreational activities and provide safety precaution information.

11

CI 5–18 · exposure 8 · augmentation 38 · importance 4.1/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Recreational instruction remains highly dependent on certified human presence, personal interaction, and real-time safety response. Adoption of AI in this domain is minimal; these sectors remain labor-intensive and resistant to automation due to safety and experiential requirements.
Sector adoption velocityclaude-sonnet-51/5Recreational/physical safety services are a low-digitization sector with minimal AI agent deployment; adoption of AI for hands-on instruction is essentially nonexistent.
Augmentation potentialclaude-haiku-4-5-202510012/5AI could marginally assist by generating pre-session content, organizing safety reminders, or providing post-session video analysis, but current systems offer limited real-time augmentation during actual instruction or emergency response scenarios.
Augmentation potentialclaude-sonnet-53/5AI can help instructors prepare lesson plans, safety scripts, or multilingual materials, and apps can supplement verbal safety instructions, offering moderate assistance without replacing the in-person role.
Task automatabilityclaude-haiku-4-5-202510011/5Teaching recreational activities and safety precautions requires real-time interaction, physical demonstration, personalized feedback, and dynamic responsiveness to individual participant needs and emergencies. Current AI cannot physically demonstrate skills, assess in-person form, or provide the embodied, adaptive instruction this task demands.
Task automatabilityclaude-sonnet-52/5Instruction requires physical demonstration, real-time observation of participant skill and safety, and hands-on correction that current AI cannot perform; only ancillary content (safety pamphlets, videos) could be automated.
Adoption barriersclaude-haiku-4-5-202510014/5Strong liability barriers exist: instructors and ski patrol are often required to be certified and legally responsible for participant safety and incident response. Regulatory requirements for swimming/ski instruction and duty-of-care standards create legal barriers to full AI substitution.
Adoption barriersclaude-sonnet-54/5Many jurisdictions require certified instructors/lifeguards for liability and safety reasons, and physical presence is needed to prevent injury or drowning, creating strong regulatory and liability barriers.
Cost vs. human wageclaude-haiku-4-5-202510011/5The human instructor is already inexpensive relative to liability costs and safety requirements. AI deployment would require significant integration, safety verification, and human oversight (especially for drowning/injury prevention), making the all-in cost exceed the marginal human wage.
Cost vs. human wageclaude-sonnet-52/5Producing generic safety videos or instructional content is cheap via AI, but the physical instruction and supervision portion still requires a human, keeping overall substitution costly relative to value delivered.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed product reliably performs live instruction of skiing, swimming, or recreational activities with safety oversight. While AI can generate instructional content or answer questions, end-to-end teaching and safety monitoring in real-world recreational settings remains research-stage and undeployed at scale.
Technical feasibility todayclaude-sonnet-51/5No deployed AI product provides in-person skiing/swimming instruction with live safety oversight; at most apps offer supplementary educational content, not the task itself.

Patrol or monitor recreational areas, such as trails, slopes, or swimming areas, on foot, in vehicles, or from towers.

8

CI 5–11 · exposure 0 · augmentation 25 · importance 4.8/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Recreational protective services operate in labor-intensive, geographically dispersed, and often small-organization contexts with slow digitization. While some resorts or facilities experiment with camera systems, meaningful displacement through automation is nascent and hindered by regulatory and safety-critical constraints.
Sector adoption velocityclaude-sonnet-51/5Recreational safety services are a low-digitization, physically-embedded sector with minimal AI agent deployment in production; adoption of even assistive tech like drowning-detection cameras remains limited and pilot-stage.
Augmentation potentialclaude-haiku-4-5-202510012/5AI-assisted tools like surveillance cameras or alert systems can help a lifeguard or patrol officer monitor larger areas, but they provide limited productivity gain since the core task—physical presence, real-time observation, and emergency intervention—remains human-dependent and cannot be significantly accelerated by current AI assistance.
Augmentation potentialclaude-sonnet-52/5Some camera-based drowning detection and monitoring tools can alert human lifeguards to potential incidents, offering modest assistance, but this is not yet widespread or central to patrol duties.
Task automatabilityclaude-haiku-4-5-202510011/5This task requires real-time physical presence, situational awareness across dynamic environments, and immediate responsiveness to emergencies that current AI systems cannot reliably replicate. Patrols demand continuous observation, spatial navigation, and split-second decision-making in uncontrolled outdoor settings where AI lacks embodied capability.
Task automatabilityclaude-sonnet-51/5Physically patrolling and monitoring recreational areas for hazards and emergencies requires embodied presence, real-time human judgment, and physical intervention capability that current AI cannot replicate end-to-end.
Adoption barriersclaude-haiku-4-5-202510014/5Strong regulatory and liability barriers exist: lifeguards and ski patrol are often legally required to be licensed, certified, and physically present; liability for failures in emergency response is severe; and public expectation and many jurisdiction laws mandate human accountability for safety in recreational areas.
Adoption barriersclaude-sonnet-54/5Liability for drowning or injury, legal requirements for certified lifeguards at many facilities, and the need for immediate physical rescue response create strong barriers to full automation.
Cost vs. human wageclaude-haiku-4-5-202510012/5Surveillance drones or automated monitoring systems are available but integration costs, human oversight requirements, and the need for on-site human responders mean total cost remains comparable to or higher than a human patrol worker in most deployments.
Cost vs. human wageclaude-sonnet-51/5Autonomous patrol systems (drones, sensor networks, camera AI) require significant capital investment, maintenance, and human oversight, and still can't replace the physical response function, making them costlier than a lifeguard's wage for equivalent coverage.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed product reliably performs autonomous patrol and monitoring of recreational areas end-to-end today. While drones and cameras exist for surveillance, they cannot legally or operationally replace the human judgment, emergency response, and safety authority required of lifeguards or ski patrol.
Technical feasibility todayclaude-sonnet-51/5No deployed product autonomously patrols trails, slopes, or pools performing the full range of human observation and physical response; camera-based drowning detection exists but is narrow and supplementary, not a replacement for patrol.

Warn recreational participants of inclement weather, unsafe areas, or illegal conduct.

5

CI 0–10 · exposure 5 · augmentation 38 · importance 4.3/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Recreational protective services are labor-intensive, locally managed, low-digitization sectors with high human-contact requirements. Adoption of AI for core warning and patrol functions is extremely limited; most organizations still rely on trained human staff for safety-critical decisions.
Sector adoption velocityclaude-sonnet-51/5Recreational safety and outdoor recreation sectors have very low AI adoption for frontline physical monitoring and intervention tasks.
Augmentation potentialclaude-haiku-4-5-202510012/5Limited augmentation potential: AI could assist with weather alerts or help flag potential drowning signatures in video, but the core task—real-time judgment and immediate warning to participants—remains inherently human. Augmentation value is narrow compared to the full task scope.
Augmentation potentialclaude-sonnet-53/5Weather apps, alert systems, and camera-based hazard detection can support situational awareness and help lifeguards decide when to warn people, but the core warning and enforcement action remains human-driven.
Task automatabilityclaude-haiku-4-5-202510011/5Detecting unsafe areas and illegal conduct requires real-time perception, judgment of context, and immediate intervention that current AI systems cannot perform end-to-end. While AI can monitor weather feeds, the task demands rapid visual assessment, behavioral judgment, and direct communication with unpredictable humans in dynamic physical environments.
Task automatabilityclaude-sonnet-51/5This requires real-time physical presence, visual scanning of people and environment, and immediate verbal/physical intervention in unpredictable outdoor settings that current AI cannot perform end-to-end.
Adoption barriersclaude-haiku-4-5-202510015/5Strong legal and liability barriers exist: most jurisdictions require licensed, accountable humans (lifeguards, ski patrol) to make safety and legal judgments, and civil/criminal liability for negligence creates hard barriers to full automation. Human presence and judgment are often legally mandated.
Adoption barriersclaude-sonnet-54/5Liability and safety regulations typically require certified human lifeguards/patrollers to be present and empowered to act, and physical intervention capability is often legally mandated.
Cost vs. human wageclaude-haiku-4-5-202510011/5Current AI infrastructure (sensors, compute, integration, real-time oversight) required for reliable deployment at a beach or ski slope would exceed the cost of employing human lifeguards or patrol staff, especially given the safety-critical nature and liability exposure of false negatives.
Cost vs. human wageclaude-sonnet-51/5There is no viable AI substitute performing this full task, so cost comparison favors the human who can physically act; sensor/camera systems would add cost without replacing the human presence requirement.
Technical feasibility todayclaude-haiku-4-5-202510012/5No deployed product performs this task reliably in production. Weather monitoring systems exist, but detecting unsafe conditions on-site and identifying illegal conduct in real-time to warn participants requires integration of computer vision, behavioral analysis, and autonomous physical presence that no fielded system accomplishes at acceptable error rates.
Technical feasibility todayclaude-sonnet-51/5No deployed product autonomously monitors beaches, ski slopes, or parks and issues real-time safety warnings to individuals; weather alert systems exist but don't cover unsafe areas or illegal conduct detection and response.

Participate in recreational demonstrations to entertain resort guests.

5

CI 0–10 · exposure 0 · augmentation 0 · importance 2.4/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5This task operates in recreational and hospitality sectors where in-person human demonstration remains the core value proposition; no adoption of AI for live entertainment demonstrations is occurring in this context.
Sector adoption velocityclaude-sonnet-51/5Hospitality/recreation sectors show low AI adoption for physical, in-person guest entertainment roles.
Augmentation potentialclaude-haiku-4-5-202510011/5AI offers minimal assistance to a human performing live recreational demonstrations. Scripting or planning tools might help marginally, but they do not meaningfully transform the core performance task.
Augmentation potentialclaude-sonnet-51/5AI offers essentially no assistance to a human physically performing recreational demonstrations for entertainment.
Task automatabilityclaude-haiku-4-5-202510011/5Recreational demonstrations require physical presence, real-time interaction with guests, spontaneous engagement, and embodied performance—capabilities that current AI systems fundamentally lack. No AI can physically perform or meaningfully substitute for human entertainers.
Task automatabilityclaude-sonnet-51/5This is a physical, in-person performance task involving human presence, skill demonstration, and entertainment value that current AI cannot physically execute.
Adoption barriersclaude-haiku-4-5-202510015/5Strong barriers exist: guest expectations require human presence and interaction, liability concerns around automated entertainment at resorts, and the inherent human-contact requirement of the task itself preclude meaningful automation.
Adoption barriersclaude-sonnet-53/5No licensing requirement specifically for entertainment demos, but the physical embodiment requirement and guest expectation of human interaction create strong practical barriers.
Cost vs. human wageclaude-haiku-4-5-202510011/5AI cannot perform this task at all, making cost comparison moot. A human lifeguard or patrol worker performing demonstrations costs substantially less per engagement than any theoretical AI substitute would.
Cost vs. human wageclaude-sonnet-51/5There is no viable AI substitute performing this physical entertainment task, so AI cost comparison is not applicable/AI is not a real option.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed product can autonomously perform recreational demonstrations for resort guests. This task requires live human performance, physical presence, and real-time audience interaction that AI systems cannot deliver.
Technical feasibility todayclaude-sonnet-51/5No deployed AI product performs physical recreational demonstrations for guests; this requires embodied human performance.

Operate underwater recovery units.

4

CI 0–9 · exposure 0 · augmentation 25 · importance 4.1/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Recreational protective services remain largely unmechanized and depend on human lifeguards. Adoption of autonomous or AI-driven underwater recovery is negligible in the sectors where this task occurs, with strong preference for trained human operators in safety-critical roles.
Sector adoption velocityclaude-sonnet-51/5This is a highly physical, low-digitization task in a sector (recreational safety/emergency response) with minimal AI adoption for hands-on physical operations.
Augmentation potentialclaude-haiku-4-5-202510012/5While sonar and underwater imaging assist human operators in locating subjects, current AI tools offer minimal augmentation to the core task of physically operating recovery units in dynamic rescue scenarios. Most augmentation remains in search and detection phases rather than active recovery operations.
Augmentation potentialclaude-sonnet-52/5AI-assisted sonar, mapping, or drone-based underwater imaging could support planning and search efforts, but the actual operation of recovery units remains manual with limited AI augmentation currently.
Task automatabilityclaude-haiku-4-5-202510011/5Underwater recovery operations require real-time environmental assessment, physical dexterity in unpredictable water conditions, and split-second decision-making that current AI systems cannot perform end-to-end. No AI system can autonomously operate underwater equipment, navigate dynamic conditions, and recover persons or objects reliably today.
Task automatabilityclaude-sonnet-51/5Operating underwater recovery units requires physical presence, manual dexterity, and real-time judgment in dynamic aquatic environments that current AI cannot perform end-to-end.atable
Adoption barriersclaude-haiku-4-5-202510015/5Recreational protective services are heavily regulated and typically require licensed, trained personnel to perform rescue operations. Legal liability, safety standards, and the requirement for human judgment in life-threatening situations create hard barriers to automation or substitution.
Adoption barriersclaude-sonnet-54/5Underwater recovery often involves certification requirements, safety protocols, and liability concerns that mandate trained human divers, though it's not always a strict legal licensing requirement like medicine or law.
Cost vs. human wageclaude-haiku-4-5-202510012/5Specialized underwater recovery equipment and ROV systems are capital-intensive and expensive to operate, often costing more than deploying trained human lifeguards for routine recreational water safety scenarios. AI-driven systems do not yet offer cost advantage for this task.
Cost vs. human wageclaude-sonnet-51/5There is no viable AI substitute performing this physical task, so AI cost per task-equivalent is effectively infinite or inapplicable compared to human labor.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed product performs autonomous underwater recovery operations in production. While remotely operated vehicles (ROVs) exist for industrial inspection, they require skilled human operators and do not function as autonomous systems for rescue recovery tasks in recreational settings.
Technical feasibility todayclaude-sonnet-51/5No deployed AI product operates underwater recovery equipment; this remains firmly in the domain of trained human divers and rescue personnel.

Contact emergency medical personnel in case of serious injury.

3

CI 0–5 · exposure 5 · augmentation 25 · importance 4.7/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Adoption of AI for autonomous emergency dispatch in recreational settings is minimal; organizations remain cautious due to liability and the requirement for accountable human judgment in life-safety decisions.
Sector adoption velocityclaude-sonnet-51/5Recreational safety and physical protective service roles are low-digitization, low-AI-adoption sectors with essentially no automation of emergency response actions.
Augmentation potentialclaude-haiku-4-5-202510012/5AI could marginally assist by alerting personnel to potential emergencies via video monitoring or sensor analysis, but the core task of assessing injury severity and contacting emergency services requires human judgment and remains the lifeguard's responsibility.
Augmentation potentialclaude-sonnet-52/5AI-enabled communication tools (e.g., automated dispatch systems, wearable sensors alerting staff) could support faster detection or notification, but the core contact/decision task itself sees minimal AI assistance today.
Task automatabilityclaude-haiku-4-5-202510011/5This task requires real-time assessment of injury severity and direct human communication with emergency services—judgments that depend on contextual understanding of medical urgency. No current AI system can autonomously determine when an injury is serious enough to warrant emergency contact or reliably communicate complex situational details to dispatchers without human oversight.
Task automatabilityclaude-sonnet-51/5This requires real-time human judgment about injury severity, immediate physical presence, and the ability to place emergency calls while managing a crisis scene; no AI system performs this end-to-end today.
Adoption barriersclaude-haiku-4-5-202510015/5This task has hard legal and liability barriers: lifeguards and ski patrol workers are often required by law or regulation to be on-site and responsible for emergency response. A human must legally assess and initiate emergency contact, and liability exposure makes autonomous AI substitution infeasible regardless of technical capability.
Adoption barriersclaude-sonnet-55/5Emergency response decisions involve life-safety liability, physical presence requirements, and often certification/training mandates (e.g., lifeguard/EMT protocols), making human authorization essential.
Cost vs. human wageclaude-haiku-4-5-202510011/5A lifeguard or ski patrol worker costs roughly $20–30k annually in loaded wages, while the integration and continuous operation of AI systems for this task (including liability oversight and fallback human confirmation) would not achieve meaningful cost savings due to the critical safety nature of the work.
Cost vs. human wageclaude-sonnet-51/5There is no viable AI substitute performing this task, so cost comparison favors the human by default since AI cannot deliver the output at all.
Technical feasibility todayclaude-haiku-4-5-202510012/5While AI systems can recognize scenarios that might suggest injury (via video analysis or sensor data), no deployed product reliably performs the full task of determining injury severity and contacting emergency services independently. Some early-stage computer vision systems exist for incident detection, but they require human confirmation before any emergency contact occurs.
Technical feasibility todayclaude-sonnet-51/5No deployed product autonomously assesses injury severity on-site and contacts EMS; this remains a human decision and action performed via phone or radio.

Provide assistance in the safe use of equipment, such as ski lifts.

3

CI 0–5 · exposure 0 · augmentation 25 · importance 3.6/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Outdoor recreation sectors are physically dispersed, operate seasonally, and involve direct human safety oversight where regulatory and cultural norms strongly favor human presence over automation.
Sector adoption velocityclaude-sonnet-51/5Outdoor recreational/physical safety services are a low-digitization sector with essentially no AI adoption for hands-on safety assistance.
Augmentation potentialclaude-haiku-4-5-202510012/5AI could assist with minor tasks like monitoring equipment status or reminding users of procedures via signage or apps, but the core task of in-person safety assistance requires human judgment and intervention that AI cannot meaningfully augment.
Augmentation potentialclaude-sonnet-52/5AI could support with sensors, cameras, or predictive maintenance alerts on equipment, but offers minimal direct assistance to the human task of physically guiding people onto lifts.
Task automatabilityclaude-haiku-4-5-202510011/5Providing equipment safety assistance requires real-time physical presence, situational judgment, and direct interaction with users in unpredictable environments. Current AI cannot perform the dynamic, embodied oversight necessary for this task.
Task automatabilityclaude-sonnet-51/5This requires physical presence to observe riders, provide hands-on assistance boarding/exiting lifts, and react instantly to safety issues; no current AI system can perform this physical guidance role.
Adoption barriersclaude-haiku-4-5-202510015/5Strong regulatory and liability barriers exist: ski areas face legal duty to maintain safe equipment operations, trained human attendants are often mandated by state/local safety codes, and liability for AI-caused injury would be catastrophic and unclear.
Adoption barriersclaude-sonnet-54/5Liability, safety regulation, and the need for immediate physical intervention in case of falls or equipment malfunction create strong barriers to automation.
Cost vs. human wageclaude-haiku-4-5-202510011/5The human cost of a trained lifeguard or ski patrol member is modest compared to the infrastructure, liability insurance, and continuous monitoring systems that would be required for AI-based alternatives to operate safely.
Cost vs. human wageclaude-sonnet-51/5There is no viable AI substitute providing the physical assistance function, so AI cost is not comparable to human labor for this task.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed AI system reliably monitors and assists users with ski lift safety or other recreational equipment in production settings. This requires embodied supervision and real-time intervention beyond current capabilities.
Technical feasibility todayclaude-sonnet-51/5No deployed AI product provides physical assistance for ski lift usage; this remains purely a human physical-presence task.

Rescue distressed persons, using rescue techniques and equipment.

0

CI 0–0 · exposure 0 · augmentation 25 · importance 4.8/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Rescue services remain highly conservative, prioritize certified human responders, and operate in sectors (parks, recreation, public safety) with slow AI adoption. The safety-critical nature and legal mandate for human operators mean adoption of autonomous rescue systems is negligible.
Sector adoption velocityclaude-sonnet-51/5Recreational safety services are a low-digitization, physically-embedded sector with minimal AI deployment for the rescue act itself, though some detection-assist pilots exist.
Augmentation potentialclaude-haiku-4-5-202510012/5AI might assist with detection (spotting distressed swimmers via vision analysis) or communication, but cannot augment the core rescue task of physically extracting and treating a person in danger. Current applications are limited to localization and alerting, not rescue execution.
Augmentation potentialclaude-sonnet-52/5AI-powered drowning detection cameras and drones can alert lifeguards faster to distress situations, providing modest assistance, but do not enhance the rescue execution itself.
Task automatabilityclaude-haiku-4-5-202510011/5Rescuing distressed persons requires real-time physical intervention in dynamic, often hazardous environments. Current AI systems cannot operate rescue equipment, swim, navigate water or terrain, or physically extract people from danger. This task is fundamentally dependent on embodied physical action that no deployed AI system can perform.
Task automatabilityclaude-sonnet-51/5This is a physical, real-time water/snow rescue requiring bodily presence, swimming, physical strength, and split-second judgment; no AI system can perform the physical rescue act itself.
Adoption barriersclaude-haiku-4-5-202510015/5Rescue operations are legally and ethically bound to trained, certified humans who bear personal and organizational liability for outcomes. Regulations explicitly require licensed lifeguards and ski patrol personnel to perform or supervise rescues; liability and safety requirements create insurmountable barriers to automation.
Adoption barriersclaude-sonnet-55/5Lifeguard/ski patrol certification is legally required, liability for failed rescues is severe, and physical human intervention is the only means of executing a rescue, creating a hard barrier to automation.
Cost vs. human wageclaude-haiku-4-5-202510011/5Any hypothetical AI rescue system (specialized robots, autonomous vehicles) would cost orders of magnitude more than the loaded wage of a lifeguard or ski patrol worker, with significant infrastructure and maintenance overhead.
Cost vs. human wageclaude-sonnet-51/5There is no AI substitute performing the physical rescue, so cost comparison is moot—human presence is mandatory and cannot be replaced by cheaper compute.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed product exists that can autonomously rescue distressed persons. While robotics research explores rescue drones and underwater robots, these remain in narrow experimental settings and cannot handle the complexity, variability, and physical demands of real rescue operations.
Technical feasibility todayclaude-sonnet-51/5No deployed product performs physical rescues; AI-assisted drone or camera detection systems exist only to flag distress, not to execute the rescue.

Examine injured persons and administer first aid or cardiopulmonary resuscitation, if necessary, using training and medical supplies and equipment.

0

CI 0–0 · exposure 0 · augmentation 25 · importance 4.7/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Recreational protective services remain predominantly human-delivered in real environments (beaches, ski slopes, pools). Organizations have not adopted AI agents to replace or conduct emergency medical assessment and treatment, and there is no measurable shift toward automation in this domain.
Sector adoption velocityclaude-sonnet-51/5Recreational safety and physical rescue services are a low-digitization, physical-presence sector with essentially no AI adoption for hands-on emergency response.
Augmentation potentialclaude-haiku-4-5-202510012/5AI might offer limited augmentation through real-time diagnostic prompts or emergency protocol reminders displayed to a human responder, but such assistance is modest. The task is inherently execution-driven and time-critical, leaving little room for AI to significantly amplify human productivity in the moment of acute intervention.
Augmentation potentialclaude-sonnet-52/5AI-enabled devices (e.g., smart AEDs with voice guidance) can offer minor procedural prompts, but they provide limited assistance to the core physical task of examining and treating an injured person.
Task automatabilityclaude-haiku-4-5-202510011/5This task requires real-time physical intervention, hands-on medical procedures (CPR chest compressions, rescue breathing, wound care), and dynamic assessment of a conscious, moving person in distress. Current AI cannot perform physical actions or deliver CPR, and remote guidance over video lacks the tactile feedback and immediate responsiveness needed for life-saving interventions.
Task automatabilityclaude-sonnet-51/5This requires real-time physical presence, hands-on medical intervention, and rapid situational judgment on an injured person, none of which current AI systems can perform end-to-end.
Adoption barriersclaude-haiku-4-5-202510015/5Lifeguarding and ski patrol are licensed and legally regulated roles; state and local law typically mandate that certified human responders assess and treat injured persons, and liability for incorrect medical decisions rests on the responsible party. Automated first aid administration would face severe legal, regulatory, and liability barriers.
Adoption barriersclaude-sonnet-55/5Administering first aid/CPR requires certified human training, physical dexterity, and legal/liability accountability, making this a hard-barrier task requiring a trained human.
Cost vs. human wageclaude-haiku-4-5-202510011/5The cost of AI systems (hardware, integration, liability insurance, redundancy) plus required human oversight cannot compete with the loaded wage of a single trained lifeguard or ski patroller. Life-critical tasks demand human presence and accountability regardless of AI cost.
Cost vs. human wageclaude-sonnet-51/5There is no AI substitute delivering this physical service, so any comparison favors the human worker entirely.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed AI system can independently examine an injured person and perform first aid or CPR. While telemedicine platforms exist for consultation, they require a human responder on-site to execute procedures; AI has no autonomous capability to deliver physical medical care in emergency settings.
Technical feasibility todayclaude-sonnet-51/5No deployed AI product examines injured people or administers physical first aid/CPR; this remains firmly outside current product capability.

Related occupations — Protective 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.