Low Vision Therapists, Orientation and Mobility Specialists, and Vision Rehabilitation Therapists

29-1122.01
Median wage $100,330/yr162,450 employed (US)Rank #884 of 923 scored · top 96% by substitution

Provide therapy to patients with visual impairments to improve their functioning in daily life activities. May train patients in activities such as computer use, communication skills, or home management skills.

Sub-scores

0–100 · band = confidence interval from rater disagreement

Substitution10
Exposure8
Augmentation38

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

21 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%10

panel mean rating 1.4/5 → substitution pressure 10/100

Technical feasibility todayw 20%6

panel mean rating 1.2/5 → substitution pressure 6/100

Cost vs. human wagew 15%9

panel mean rating 1.4/5 → substitution pressure 9/100

Adoption barriersw 20%inverted — strong barriers lower the score17

panel mean rating 4.3/5 (barrier strength) → substitution pressure 17/100

Sector adoption velocityw 10%5

panel mean rating 1.2/5 → substitution pressure 5/100

Task breakdown (21 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.

Write reports or complete forms to document assessments, training, progress, or follow-up outcomes.

70

CI 60–80 · exposure 78 · augmentation 88 · importance 4.4/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510013/5Healthcare sectors show middling adoption of AI documentation tools; many clinical practices pilot such systems but full production deployment remains inconsistent, particularly in specialized rehabilitation settings.
Sector adoption velocityclaude-sonnet-52/5Allied health and vision rehabilitation is a small, specialized, and less digitized sector where AI documentation tools have seen limited penetration compared to general medicine or corporate settings.
Augmentation potentialclaude-haiku-4-5-202510015/5AI excels at generating first drafts, organizing assessment data, and standardizing language while clinicians review and customize, significantly boosting documentation productivity while maintaining clinician authority and responsibility.
Augmentation potentialclaude-sonnet-54/5AI can meaningfully speed up drafting of assessment summaries and progress notes from therapist-provided input, letting the specialist focus on review and clinical judgment rather than writing from scratch.
Task automatabilityclaude-haiku-4-5-202510015/5Documentation and form completion following structured assessments are highly automatable; current AI systems can reliably extract assessment data, generate comprehensive reports from clinical notes, and populate standardized forms with minimal human input, easily meeting the 50% time-saving threshold.
Task automatabilityclaude-sonnet-54/5Report writing and form completion from structured clinical observations is largely language generation from notes, a task well-suited to current LLMs with human review, though clinical nuance requires oversight.
Adoption barriersclaude-haiku-4-5-202510013/5Professional licensing requires clinician oversight and sign-off on reports documenting patient progress, and liability concerns around clinical accuracy create meaningful but not prohibitive barriers; generated reports still need human review and authorization.
Adoption barriersclaude-sonnet-53/5Clinical documentation often requires the credentialed therapist's sign-off for accuracy, liability, and reimbursement purposes, creating moderate barriers despite no outright prohibition on AI drafting.
Cost vs. human wageclaude-haiku-4-5-202510015/5AI-driven documentation generation costs far less than clinician time spent on form-filling and report writing; inference and oversight are inexpensive compared to professional wages, yielding an order-of-magnitude cost advantage.
Cost vs. human wageclaude-sonnet-54/5AI-assisted drafting from dictated or typed notes is far cheaper per report than a specialist's full documentation time, even after factoring in review and correction.
Technical feasibility todayclaude-haiku-4-5-202510014/5Mature products exist (EHR systems with templated documentation, AI-assisted clinical note generation) and are deployed in healthcare settings, though integration with specialized vision rehabilitation workflows and final clinical sign-off requirements introduce minor friction.
Technical feasibility todayclaude-sonnet-53/5AI dictation and clinical note-generation tools are deployed in some healthcare/therapy settings, but vision rehabilitation-specific documentation templates and terminology are less commonly integrated into mature commercial products.

Participate in professional development activities, such as reading literature, continuing education, attending conferences, and collaborating with colleagues.

26

CI 16–35 · exposure 17 · augmentation 63 · importance 4.0/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Adoption of AI in professional development for healthcare occupations remains very early; while some organizations use AI tools to recommend content or organize resources, actual displacement of human participation in conferences, reading, and peer collaboration is minimal and unlikely.
Sector adoption velocityclaude-sonnet-52/5Healthcare and allied health rehabilitation fields adopt AI tools slowly overall, though individual practitioners increasingly use AI search/summarization tools for continuing education informally.
Augmentation potentialclaude-haiku-4-5-202510013/5AI can assist by summarizing literature, recommending relevant continuing education courses, organizing conference schedules, or transcribing notes from colleague discussions, offering meaningful productivity gains in information curation while the human remains fully in control of learning decisions.
Augmentation potentialclaude-sonnet-54/5AI can meaningfully assist by summarizing research literature, recommending relevant continuing education content, and helping synthesize conference takeaways, boosting efficiency while the human remains the active participant.
Task automatabilityclaude-haiku-4-5-202510011/5This task fundamentally requires human judgment about professional growth, selection of learning materials aligned with individual expertise gaps, and collaborative exchange of ideas with peers. AI cannot autonomously decide what a therapist should learn or generate the reflective, personalized benefit from conferences and colleague interactions.
Task automatabilityclaude-sonnet-52/5AI can help summarize literature or surface relevant articles, but 'participating' in professional development including attending conferences and collaborating with colleagues inherently requires human presence and engagement that cannot be fully automated.dataThe task is largely a human activity mediated by AI tools rather than replaceable.
Adoption barriersclaude-haiku-4-5-202510014/5Professional development is intrinsically tied to licensure, credential maintenance, and board requirements in most jurisdictions. Therapists are often legally required to demonstrate active continuing education and professional engagement, which must involve human participation and cannot be delegated to AI.
Adoption barriersclaude-sonnet-52/5No strict licensing barrier prevents AI assistance, but professional development is largely a personal/social requirement often tied to certification maintenance, creating some structural expectation of human participation.
Cost vs. human wageclaude-haiku-4-5-202510012/5AI-assisted tools (e.g., literature summarization) reduce time spent on information gathering, but the core task—active participation, reflection, and peer collaboration—cannot be offloaded. The cost of human time in professional development typically far exceeds marginal AI cost savings on information processing.
Cost vs. human wageclaude-sonnet-52/5AI tools for literature summarization are cheap, but the bulk of the task's value (networking, live discussion, hands-on collaboration) has no AI substitute, so overall cost comparison favors humans doing the core activity themselves.
Technical feasibility todayclaude-haiku-4-5-202510012/5While AI can summarize literature, curate conference materials, or facilitate scheduling of collaboration, no deployed product autonomously manages a therapist's professional development in a way that replaces human participation in reading, learning, and peer engagement. This remains supervisory or preparatory work.
Technical feasibility todayclaude-sonnet-52/5Products like AI summarizers and research assistants exist and are used to digest literature, but no deployed product performs the full scope of professional development participation (networking, live conference engagement, peer collaboration).

Provide consultation, support, or education to groups such as parents and teachers.

15

CI 0–30 · exposure 13 · augmentation 50 · importance 4.2/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Vision rehabilitation services operate in small, specialized, often non-profit and clinical settings with low digitization and strong professional norms favoring direct human contact; adoption of AI agents for consultation is minimal.
Sector adoption velocityclaude-sonnet-52/5Vision rehabilitation and specialized therapy services are a niche, relationship-driven field with low digitization and slow AI adoption compared to sectors like finance or generic professional services.
Augmentation potentialclaude-haiku-4-5-202510012/5AI might assist with preparing educational materials or structuring talking points before a consultation, but the core task—live group consultation and adaptive support—resists meaningful augmentation from current systems.
Augmentation potentialclaude-sonnet-54/5AI can meaningfully assist by drafting educational handouts, generating FAQs, translating materials, or summarizing best practices for parents and teachers, boosting the specialist's efficiency while they retain direct interaction.
Task automatabilityclaude-haiku-4-5-202510011/5This task requires nuanced, empathetic human interaction and adaptation to group dynamics. Current AI cannot replicate the relational and contextual judgment needed to consult effectively with parents and teachers in a rehabilitation setting.
Task automatabilityclaude-sonnet-52/5Delivering group consultation and education involves live interaction, reading audience needs, and adapting explanations about vision rehabilitation, which AI cannot fully replace end-to-end today, though it can help prepare materials.
Adoption barriersclaude-haiku-4-5-202510015/5This task involves direct service delivery to vulnerable populations (parents of children with vision loss, educational staff) and likely requires licensing or certification by the therapist; regulatory and liability frameworks strongly protect human-delivered consultation and support.
Adoption barriersclaude-sonnet-53/5While no strict licensing law bars AI from generating educational content, the sensitive, personalized nature of consulting with parents/teachers about a child's or client's vision impairment creates strong preference and trust barriers favoring human specialists.
Cost vs. human wageclaude-haiku-4-5-202510011/5A specialist vision rehabilitation therapist commanding a professional salary cannot be cost-effectively replaced by AI for group consultation work, which demands credentialing, accountability, and presence.
Cost vs. human wageclaude-sonnet-52/5Since a human specialist must still deliver the actual consultation and answer nuanced questions, AI only reduces prep costs marginally rather than replacing the labor cost of live delivery.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed product reliably delivers consultation and support to live groups of parents and teachers in a professional rehabilitation context; this remains outside the scope of production AI systems.
Technical feasibility todayclaude-sonnet-52/5AI tools can generate educational content or slides, but no deployed product independently conducts consultation sessions with parents/teachers on vision rehabilitation topics reliably in production.

Design instructional programs to improve communication, using devices such as slates and styluses, braillers, keyboards, adaptive handwriting devices, talking book machines, digital books, and optical character readers (OCRs).

13

CI 4–23 · exposure 13 · augmentation 50 · importance 4.1/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Vision rehabilitation is a specialized, low-digitization healthcare sector with small teams, primarily in clinical/nonprofit settings with limited tech infrastructure and strong professional licensing requirements that slow automation adoption.
Sector adoption velocityclaude-sonnet-51/5Vision rehabilitation services are a small, specialized allied health field with low digitization and minimal reported AI adoption in program design workflows.
Augmentation potentialclaude-haiku-4-5-202510013/5AI can usefully assist by researching device features, generating draft instructional outlines, or organizing evidence-based content templates, but the core task of designing patient-centered programs still depends on the therapist's clinical expertise and iterative refinement.
Augmentation potentialclaude-sonnet-53/5AI tools can help therapists research assistive technology options, draft lesson plans, and generate materials for adaptive devices, offering moderate productivity support alongside required clinical judgment.
Task automatabilityclaude-haiku-4-5-202510011/5Designing instructional programs requires individualized assessment of patient needs, understanding of disabilities, pedagogical expertise, and iterative program customization. These demand human clinical judgment and personalized adaptation that current AI cannot perform end-to-end with quality parity.
Task automatabilityclaude-sonnet-52/5Designing individualized instructional programs requires assessing a specific client's residual vision, cognitive ability, and daily needs, which current AI cannot autonomously evaluate or synthesize into a full program; AI can assist with drafting materials but not replace the end-to-end clinical design process.
Adoption barriersclaude-haiku-4-5-202510015/5This task requires licensed rehabilitation professionals (many jurisdictions require certification/licensure) and involves direct responsibility for patient learning outcomes, creating legal and regulatory barriers that mandate human professional judgment and accountability.
Adoption barriersclaude-sonnet-54/5This work typically requires certified professionals (COMS, CLVT credentials) and involves direct client assessment and safety-critical skill-building, creating strong professional and liability barriers to full automation.
Cost vs. human wageclaude-haiku-4-5-202510012/5AI tools for content drafting and research are inexpensive, but the integration overhead, clinical oversight, and the need for human specialist validation mean the total cost remains comparable to or potentially exceeds a therapist's time on this complex design task.
Cost vs. human wageclaude-sonnet-52/5While AI-assisted content generation is cheap, the actual value-add—clinical assessment and personalized program design—still requires a licensed specialist, so all-in cost savings are limited.
Technical feasibility todayclaude-haiku-4-5-202510012/5While AI can assist with content generation and device information retrieval, no deployed product reliably designs comprehensive rehabilitation instructional programs that integrate device selection, learning objectives, and patient-specific accessibility needs at production scale.
Technical feasibility todayclaude-sonnet-51/5There are no deployed AI products that autonomously create individualized vision rehabilitation instructional programs; this remains a specialist clinical task performed by trained therapists.

Develop rehabilitation or instructional plans collaboratively with clients, based on results of assessments, needs, and goals.

9

CI 0–18 · exposure 8 · augmentation 50 · importance 4.5/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Vision rehabilitation is a clinical specialty with low overall digitization and strong regulatory/professional oversight; adoption of AI for core clinical planning tasks has been minimal and remains highly cautious.
Sector adoption velocityclaude-sonnet-51/5Vision rehabilitation and allied health/education services are a small, low-digitization sector with minimal AI agent deployment for care planning.
Augmentation potentialclaude-haiku-4-5-202510013/5AI could assist therapists by summarizing assessment data, suggesting evidence-based intervention options, or helping organize client goals, but the collaboration and personalization must remain under therapist direction and judgment.
Augmentation potentialclaude-sonnet-53/5AI can help organize assessment data, suggest goal frameworks, and draft plan documentation, giving moderate productivity benefits while the therapist retains full client interaction and decision-making.
Task automatabilityclaude-haiku-4-5-202510011/5Creating individualized rehabilitation plans requires deep understanding of client-specific assessments, personal goals, values, and life context—factors that demand human judgment and empathy. Current AI systems cannot reliably synthesize complex clinical data with client preferences to produce clinically sound, personalized care plans.
Task automatabilityclaude-sonnet-52/5Drafting a plan template can be aided by AI, but genuine collaborative goal-setting with a client requires in-person rapport, physical assessment interpretation, and negotiation of personal goals that current AI cannot autonomously conduct end-to-end.
Adoption barriersclaude-haiku-4-5-202510015/5Vision rehabilitation planning is legally required to be performed by or under the direct supervision of a licensed rehabilitation specialist; client safety, liability, and regulatory requirements create hard barriers to unsupervised AI automation.
Adoption barriersclaude-sonnet-54/5This is a licensed clinical/therapeutic task often requiring credentialed professionals and documented client-specific planning, with liability and professional standards limiting full automation.
Cost vs. human wageclaude-haiku-4-5-202510011/5The task requires licensed, specialized therapists whose expertise commands substantial wages; AI systems cannot yet perform this work reliably enough to justify substitution, and oversight costs would likely exceed the savings from automation attempts.
Cost vs. human wageclaude-sonnet-52/5While AI drafting tools are cheap, the actual value-add—client interaction, assessment synthesis, and rapport-based negotiation—still requires the specialist's time, so overall cost savings are modest.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed products reliably perform individualized rehabilitation plan development for vision-impaired clients in production settings. Such plans require integration of clinical assessment results with client collaboration and ongoing adjustment, which remains beyond current AI capabilities in healthcare.
Technical feasibility todayclaude-sonnet-51/5No deployed product performs collaborative rehabilitation planning for vision-impaired clients in production; this remains a human clinical-judgment task with no mature AI substitute in the field.

Administer tests and interpret test results to develop rehabilitation plans for clients.

9

CI 0–18 · exposure 8 · augmentation 38 · importance 3.8/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5This is a small, specialized clinical field with strong human-centric and regulatory requirements. Adoption of automation in vision rehabilitation is minimal; practitioners continue to deliver in-person, individualized assessment and planning.
Sector adoption velocityclaude-sonnet-51/5This is a small, highly specialized allied health field with low digitization and minimal reported AI deployment for functional assessment tasks.
Augmentation potentialclaude-haiku-4-5-202510012/5AI might assist with organizing or analyzing test data and generating preliminary documentation, but the core assessment and clinical judgment remain the therapist's responsibility, and such assistance offers modest productivity gains in a task driven by direct client interaction and professional expertise.
Augmentation potentialclaude-sonnet-53/5AI can help organize test data, draft documentation, and suggest evidence-based intervention options, moderately aiding therapists' planning and reporting workflow.
Task automatabilityclaude-haiku-4-5-202510011/5This task requires individualized assessment, clinical judgment, and direct interaction with clients to understand their specific vision loss and functional needs. Current AI systems cannot reliably administer standardized vision rehabilitation tests or interpret complex clinical results in the context of a patient's unique circumstances without human oversight.
Task automatabilityclaude-sonnet-52/5Test administration requires hands-on assessment of functional vision and physical mobility skills, and interpretation for rehabilitation planning requires clinical judgment integrating client-specific context that current AI cannot fully replicate end-to-end.'
Adoption barriersclaude-haiku-4-5-202510015/5Vision rehabilitation therapy is a regulated, licensed profession; only credentialed practitioners (OTRs, VRTs, O&M specialists) can legally administer standardized tests and develop clinical rehabilitation plans. Liability, professional licensing, and legal requirements create hard barriers to automation.
Adoption barriersclaude-sonnet-54/5Rehabilitation plans typically require certified/licensed specialists to administer tests and sign off on clinical interpretations, creating strong regulatory and liability barriers to automation.
Cost vs. human wageclaude-haiku-4-5-202510011/5The expertise required—clinical assessment, test administration, and personalized rehabilitation plan development—commands significant human labor cost. Current AI cannot perform these functions independently, making any application additive rather than substitutive in cost.
Cost vs. human wageclaude-sonnet-52/5Because AI cannot independently perform the physical testing or produce liability-safe rehab plans, any AI use requires substantial human oversight, keeping costs comparable to or only marginally below human-only delivery.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed products perform end-to-end test administration and clinical interpretation for vision rehabilitation planning. While AI can assist with data analysis, the medical and therapeutic assessment components require licensed human practitioners in real clinical settings.
Technical feasibility todayclaude-sonnet-51/5No deployed product administers vision/mobility functional assessments or generates clinically validated rehab plans in production; this remains outside current commercial AI offerings for this specialized domain.

Train clients to read or write Braille.

9

CI 0–18 · exposure 8 · augmentation 38 · importance 3.7/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Vision rehabilitation is a small, highly regulated, human-dependent sector with minimal digitization; adoption of AI in specialized therapeutic tasks remains negligible and client preference strongly favors direct human instruction.
Sector adoption velocityclaude-sonnet-51/5Vision rehabilitation is a small, specialized, low-digitization field with little evidence of AI agents being deployed to replace hands-on instructors.
Augmentation potentialclaude-haiku-4-5-202510012/5AI might assist with supplementary content (e.g., generating practice materials or tracking progress documentation), but offers minimal productivity uplift for the core tactile, adaptive teaching that defines Braille instruction.
Augmentation potentialclaude-sonnet-53/5AI can help generate Braille materials, track progress, or provide supplementary practice exercises, offering moderate assistance to therapists without replacing the core teaching interaction.
Task automatabilityclaude-haiku-4-5-202510011/5Braille instruction requires adaptive physical demonstration, real-time tactile feedback adjustment, and personalized pedagogical response to individual learning pace and cognitive needs—capabilities that current AI systems cannot meaningfully automate end-to-end or with 50% time savings at equal quality.
Task automatabilityclaude-sonnet-52/5Braille instruction involves hands-on tactile guidance, physical demonstration, and adaptive real-time feedback on a client's finger positioning and reading technique that current AI cannot perform end-to-end.'
Adoption barriersclaude-haiku-4-5-202510015/5Braille instruction is regulated within rehabilitation and special education frameworks; therapists must be certified, and the task inherently requires licensed human oversight and direct tactile interaction with vulnerable clients—creating legal and professional barriers to automation.
Adoption barriersclaude-sonnet-54/5This work typically requires certified vision rehabilitation professionals, involves close physical/tactile interaction, and client trust and individualized pacing create strong barriers to full automation.
Cost vs. human wageclaude-haiku-4-5-202510011/5Braille training is labor-intensive and highly personalized, requiring skilled human instructors; AI systems offer no cost advantage and cannot replace the specialized training, expertise, and one-on-one engagement needed.
Cost vs. human wageclaude-sonnet-52/5Any AI-based supplementary tools still require a human therapist for the bulk of hands-on instruction, so cost savings versus a full human specialist are minimal to moderate at best.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed AI product reliably performs Braille instruction in production; the task demands live tactile guidance, hand-over-hand correction, and dynamic assessment of a client's sensory processing that exceed current capabilities.
Technical feasibility todayclaude-sonnet-51/5No deployed AI product independently teaches Braille reading/writing to visually impaired clients in production; existing tools are limited to translation or practice supplements, not instruction delivery.

Train clients to use adaptive equipment, such as large print, reading stands, lamps, writing implements, software, and electronic devices.

8

CI 0–16 · exposure 8 · augmentation 38 · importance 4.0/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5This is a specialized, small-scale healthcare and rehabilitation sector with limited digitization and slow technology adoption. Demand remains driven by in-person service delivery, and there is minimal evidence of AI-driven displacement or automation pilots in production within vision rehabilitation clinics.
Sector adoption velocityclaude-sonnet-51/5Vision rehabilitation is a small, specialized, in-person healthcare-adjacent field with low digitization and minimal AI agent deployment in production settings.
Augmentation potentialclaude-haiku-4-5-202510012/5AI could assist by generating printed instructional guides or video tutorials for equipment use, but the core value of training—demonstrating proper technique, correcting form, and building confidence through hands-on practice—remains fundamentally human-centric. Marginal augmentation potential exists for pre-training or reference materials.
Augmentation potentialclaude-sonnet-53/5AI can help create training materials, generate step-by-step guides, recommend equipment, and provide software tutorials, but the core hands-on physical training still relies on the human specialist.
Task automatabilityclaude-haiku-4-5-202510011/5This task requires personalized assessment of individual clients' vision capabilities, hands-on demonstration, real-time feedback, and behavioral reinforcement—activities that demand human presence, judgment, and adaptive coaching. Current AI systems lack the embodied interaction, tactile guidance, and individualized safety assessment this training demands.
Task automatabilityclaude-sonnet-52/5This requires hands-on physical demonstration, personalized adjustment for individual vision impairments, and real-time physical guidance that current AI cannot perform end-to-end, though some instructional content generation could be assisted.'
Adoption barriersclaude-haiku-4-5-202510015/5Vision rehabilitation and orientation-and-mobility training are delivered by licensed or regulated practitioners in most jurisdictions. Clients require direct human contact for safety reasons (physical guidance, spatial awareness), and liability for errors in adaptive equipment training creates strong legal barriers to full automation.
Adoption barriersclaude-sonnet-54/5This work is often performed by credentialed rehabilitation specialists with certification requirements, involves vulnerable populations (visually impaired clients), and requires physical, hands-on trust-based interaction that strongly favors human delivery.
Cost vs. human wageclaude-haiku-4-5-202510011/5The specialized expertise of a licensed rehabilitation therapist commands significant loaded wages ($50–80k+ annually), and effective training requires sustained one-on-one interaction that would require multiple human trainers or a hybrid model to match. AI-only solutions cannot yet justify cost parity or advantage for personalized mobility and equipment training.
Cost vs. human wageclaude-sonnet-51/5AI cannot replace the in-person, hands-on instruction and physical guidance needed, so a human specialist remains necessary and cheaper than any AI-plus-human hybrid attempting the same outcome.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed AI product performs hands-on adaptive equipment training for low-vision clients at scale. While AI can provide instructional content or device tutorials, it cannot replicate the personalized, in-person assessment and corrective coaching that defines this clinical rehabilitation role.
Technical feasibility todayclaude-sonnet-51/5No deployed AI product currently trains low-vision clients on physical adaptive equipment use in person; this remains a human-delivered clinical/therapeutic service.

Refer clients to services, such as eye care, health care, rehabilitation, and counseling, to enhance visual and life functioning or when condition exceeds scope of practice.

8

CI 0–16 · exposure 8 · augmentation 50 · importance 3.6/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Vision rehabilitation services remain in lower-digitization sectors with significant human-contact requirements and small organizational structures; adoption of AI for autonomous referral decisions is minimal because the professional responsibility and client vulnerability make automation impractical.
Sector adoption velocityclaude-sonnet-52/5Vision rehabilitation and allied health services are a smaller, less digitized sector with limited AI agent deployment for care coordination tasks.
Augmentation potentialclaude-haiku-4-5-202510013/5AI could usefully assist by providing searchable databases of available services, summarizing client histories, or suggesting referral options for the therapist to review; however, the therapist must retain final decision-making authority, and current augmentation tools are limited to information support rather than transforming productivity significantly.
Augmentation potentialclaude-sonnet-53/5AI can help therapists search for local resources, draft referral letters, and organize client information, but the core judgment of when and where to refer remains human-led.
Task automatabilityclaude-haiku-4-5-202510011/5This task requires professional judgment about client needs, understanding of scope of practice limitations, knowledge of available services, and relationship-building with the client—all human-centered and contextual work that current AI cannot reliably perform end-to-end. Referral decisions depend on nuanced assessment of individual circumstances and treatment trajectories that exceed pattern matching.
Task automatabilityclaude-sonnet-52/5Referral decisions require clinical judgment about client needs, scope-of-practice boundaries, and coordination with local resources, which AI cannot reliably assess end-to-end today.
Adoption barriersclaude-haiku-4-5-202510015/5This task is protected by licensing requirements (therapists must be credentialed and legally responsible for clinical decisions), liability asymmetry (a poor referral decision harms a vulnerable client), and the requirement that a human professional sign off on referral decisions as part of their scope of practice and duty of care.
Adoption barriersclaude-sonnet-54/5Referral and scope-of-practice determinations are tied to professional licensure and liability, since therapists must recognize when a condition requires another licensed provider.
Cost vs. human wageclaude-haiku-4-5-202510011/5The human cost is embedded in the licensed therapist's time; deploying AI to replace this task would require significant oversight, liability management, and validation, making it more expensive than the therapist simply making the referral themselves as part of standard clinical practice.
Cost vs. human wageclaude-sonnet-51/5Referral decisions require professional liability and judgment that AI cannot substitute for, so human cost is unavoidable regardless of any AI assistance layered on top.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed product reliably performs this task autonomously; it fundamentally requires a licensed professional to assess the client's condition, determine whether a referral is appropriate, and match them to suitable services based on clinical judgment and professional networks. AI tools may assist in information lookup, but the referral decision itself remains undeployed as an autonomous AI function.
Technical feasibility todayclaude-sonnet-51/5No deployed product autonomously identifies when a vision rehabilitation case exceeds scope and executes appropriate referrals; this remains a human clinical judgment task.

Identify visual impairments related to basic life skills in areas such as self care, literacy, communication, health management, home management, and meal preparation.

7

CI 5–10 · exposure 5 · augmentation 25 · importance 4.1/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Vision rehabilitation is a small, specialized healthcare field with low digital adoption rates and strong reliance on in-person clinical assessment. Therapists work in rehabilitation centers, clinics, and homes—settings with limited tech infrastructure and high preference for licensed human practitioners.
Sector adoption velocityclaude-sonnet-51/5This is a small, highly specialized allied health field with low digitization and no significant reported AI adoption for hands-on functional assessments.
Augmentation potentialclaude-haiku-4-5-202510012/5AI could assist by analyzing patient questionnaire responses or organizing assessment data, but the core task—clinical observation and diagnostic reasoning across multiple life domains—remains best performed by a trained human. Modest augmentation potential in documentation and data synthesis, but limited transformation of the core clinical function.
Augmentation potentialclaude-sonnet-52/5AI could help with documentation, checklists, or referencing assessment criteria, but offers limited assistance for the core hands-on evaluative work.
Task automatabilityclaude-haiku-4-5-202510011/5This task requires comprehensive clinical assessment and real-time observation of a patient performing activities in varied environments, with careful judgment about which specific visual impairments are limiting each life skill. Current AI systems cannot reliably observe, assess, and diagnose visual impairments across the full range of real-world conditions and functional contexts required here.
Task automatabilityclaude-sonnet-51/5This requires in-person functional vision assessment combined with clinical judgment about how impairments affect daily living, which cannot be performed end-to-end by current AI systems.
Adoption barriersclaude-haiku-4-5-202510014/5Vision rehabilitation assessment is typically performed by licensed professionals (therapists, optometrists, or rehabilitation specialists) whose credentials and clinical judgment are often legally or organizationally required. Liability and safety concerns around misdiagnosis of visual impairments create strong barriers to full automation.
Adoption barriersclaude-sonnet-54/5This task typically requires a credentialed rehabilitation professional to conduct assessments and make clinical determinations affecting client care plans, creating strong professional and liability barriers.
Cost vs. human wageclaude-haiku-4-5-202510011/5The cost of integrating and maintaining AI assessment tools, combined with required human oversight and validation, exceeds the loaded wage of a single therapist who can perform this diagnostic evaluation directly with no AI support.
Cost vs. human wageclaude-sonnet-51/5AI cannot substitute for the physical, interactive assessment process, so there is no viable cost comparison—human specialists remain necessary.
Technical feasibility todayclaude-haiku-4-5-202510012/5While AI can assist with some screening questionnaires or analyze video clips of visual function, no deployed product reliably performs comprehensive identification of visual impairments across multiple life domains as a therapist does. Existing systems lack the clinical judgment and environmental adaptability needed for production use in rehabilitation settings.
Technical feasibility todayclaude-sonnet-51/5No deployed product performs hands-on functional visual impairment assessments tied to life-skills domains; this remains a specialist clinical evaluation task.

Recommend appropriate mobility devices or systems, such as human guides, dog guides, long canes, electronic travel aids (ETAs), and other adaptive mobility devices (AMDs).

7

CI 0–14 · exposure 8 · augmentation 25 · importance 4.7/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Vision rehabilitation services operate in low-digitization, regulated healthcare settings with strong emphasis on human contact and clinical oversight. Adoption of autonomous AI in this space is minimal; therapy remains human-centered.
Sector adoption velocityclaude-sonnet-51/5This is a small, highly specialized, in-person allied health field with minimal AI tool adoption reported; it is not part of fast-digitizing sectors like finance or general professional services.
Augmentation potentialclaude-haiku-4-5-202510012/5AI could provide limited assistance by organizing device information or flagging relevant options for the therapist to review, but the core task of matching devices to client needs remains primarily a human clinical function where AI adds marginal value.
Augmentation potentialclaude-sonnet-52/5AI could help organize client history, summarize assessment notes, or suggest general device options for review, but it does not materially transform the core judgment-based recommendation process.
Task automatabilityclaude-haiku-4-5-202510011/5This task requires understanding individual client capabilities, limitations, preferences, lifestyle, and environmental context—nuanced clinical judgment that current AI cannot perform end-to-end. While AI might assist in information retrieval about device options, the core recommendation demands hands-on assessment, personalized matching, and real-time adjustment that remains beyond current system capability.
Task automatabilityclaude-sonnet-52/5Recommending mobility devices requires hands-on functional vision assessment, physical environment evaluation, and individualized clinical judgment that current AI cannot perform end-to-end, though AI could assist with information gathering or documentation.el.
Adoption barriersclaude-haiku-4-5-202510015/5Strong regulatory and professional barriers exist: this task involves medical assessment and rehabilitation planning that typically requires state licensure and professional credentials. Liability and duty-of-care obligations mean a licensed human must evaluate the client and bear responsibility for the recommendation.
Adoption barriersclaude-sonnet-54/5O&M specialists and vision rehab therapists are often certified/licensed professionals whose recommendations affect client safety (e.g., cane vs. guide dog vs. ETA), creating strong liability and professional-standard barriers to automation.
Cost vs. human wageclaude-haiku-4-5-202510011/5The cost of an AI system capable of reliable clinical assessment and recommendation, including integration, oversight, and liability management, far exceeds the labor cost of a therapist performing this task, especially when factoring in the clinical liability of errors.
Cost vs. human wageclaude-sonnet-51/5The task requires in-person assessment and physical trials with devices, so AI cannot substitute for the human specialist's cost at all; any AI tool would be additive, not a cheaper replacement.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed product reliably performs the full recommendation task independently. Systems exist to provide general information about mobility devices, but recommending the appropriate device for a specific vision-impaired individual requires clinical assessment and human expertise not yet automated at production scale in healthcare.
Technical feasibility todayclaude-sonnet-51/5No deployed product independently assesses a client's residual vision, gait, cognitive ability, and environment to make credentialed mobility device recommendations; this remains a specialist clinical judgment task.

Obtain, distribute, or maintain low vision devices.

6

CI 0–13 · exposure 0 · augmentation 25 · importance 3.9/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Vision rehabilitation services are delivered in small clinical and nonprofit settings with limited digitization and heavy reliance on in-person, hands-on work. Adoption of automation in this sector is minimal, as the task is not amenable to remote or digital substitution.
Sector adoption velocityclaude-sonnet-52/5Healthcare and rehabilitation services sectors are generally slower adopters of AI for hands-on physical tasks, with adoption concentrated in administrative rather than device-handling functions.
Augmentation potentialclaude-haiku-4-5-202510012/5While digital tools could assist with inventory tracking or educational materials, the core task—physical distribution and maintenance of devices—receives minimal productivity gain from current AI. Assistive potential is limited to administrative support at the margins.
Augmentation potentialclaude-sonnet-52/5AI could assist with inventory tracking, ordering reminders, or device recommendation research, but offers limited assistance for the physical distribution and fitting components.
Task automatabilityclaude-haiku-4-5-202510011/5This task requires hands-on physical distribution and maintenance of specialized devices, coupled with assessment of individual patient needs and device fit—work that fundamentally depends on in-person interaction and tactile adjustment. Current AI systems cannot physically handle, adjust, or deliver devices to patients.
Task automatabilityclaude-sonnet-51/5This task requires physical procurement, handling, fitting, and distribution of assistive devices to clients, which cannot be performed end-to-end by current AI systems.
Adoption barriersclaude-haiku-4-5-202510015/5This task is protected by strong barriers: vision rehabilitation therapists are state-licensed healthcare professionals in most jurisdictions, and device fitting and maintenance inherently require licensed human judgment, physical presence, and liability accountability. Regulatory and licensure requirements mandate human professional involvement.
Adoption barriersclaude-sonnet-53/5While not strictly licensed work, distributing medical/assistive devices typically involves professional judgment, client fitting, and organizational procurement processes that create moderate friction against automation.
Cost vs. human wageclaude-haiku-4-5-202510011/5AI has no role in the core physical and interpersonal work of obtaining, distributing, and maintaining devices; there is no cost comparison because automation does not meaningfully occur today.
Cost vs. human wageclaude-sonnet-51/5AI cannot perform the physical logistics and client interaction involved, so there is no viable AI cost comparison; a human must still execute this task.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed AI product performs the physical inventory management, device distribution, or maintenance aspects of this task reliably in production. This is not a digitized, software-only workflow.
Technical feasibility todayclaude-sonnet-51/5No deployed AI product manages physical inventory, procurement, or hands-on distribution of low vision devices to patients.

Train clients with visual impairments to use mobility devices or systems, such as human guides, dog guides, electronic travel aids (ETAs), and other adaptive mobility devices (AMDs).

3

CI 0–5 · exposure 0 · augmentation 25 · importance 4.5/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5This is a specialized clinical service in the health and rehabilitation sector, performed by licensed professionals in clinical, educational, and community settings. Adoption of AI automation is not occurring because the task is legally and ethically bound to human practitioners.
Sector adoption velocityclaude-sonnet-51/5Rehabilitation and disability services is a small, specialized, hands-on sector with minimal AI agent deployment or production-level automation of physical mobility training.
Augmentation potentialclaude-haiku-4-5-202510012/5AI could assist with scheduling, data logging of client progress, or provision of supplemental educational materials about mobility devices, but current systems offer minimal direct assistance to the core act of training clients in real-world mobility techniques.
Augmentation potentialclaude-sonnet-52/5AI could help with generating training materials, tracking progress notes, or providing supplementary app-based orientation cues, but it plays a minor role in the core physical instruction process.
Task automatabilityclaude-haiku-4-5-202510011/5This task requires sustained, real-time physical interaction, spatial judgment, safety assessment, and adaptive responsiveness to individual client needs in unpredictable environments. Current AI systems cannot safely guide a visually impaired person through variable terrain or teach embodied mobility skills without a physical human presence.
Task automatabilityclaude-sonnet-51/5This requires hands-on, physical training with real-time observation of client movement, balance, spatial awareness, and safety in real-world environments (streets, traffic, obstacles) that current AI cannot perform or supervise.
Adoption barriersclaude-haiku-4-5-202510015/5This task is heavily regulated and licensure-protected in most jurisdictions; practitioners must be certified or licensed rehabilitation specialists. Legal, liability, and ethical requirements mandate that a qualified human therapist directly oversee and perform mobility training for safety-critical reasons.
Adoption barriersclaude-sonnet-54/5Orientation and mobility specialists typically require certification/licensure, and safety-critical training (e.g., street crossings) creates strong liability concerns that favor a qualified human trainer being physically present.
Cost vs. human wageclaude-haiku-4-5-202510011/5AI solutions that might assist with theoretical instruction or simulation cannot replace the core service delivery of in-person mobility training. The cost of developing and deploying such systems would far exceed the cost of human therapist time for this safety-critical, hands-on task.
Cost vs. human wageclaude-sonnet-51/5There is no AI substitute performing this physical instruction task, so cost comparison favors the human specialist entirely; AI cannot generate the required output at all.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed AI system can independently train a visually impaired person to use mobility devices in real-world settings. This requires physical guidance, safety oversight, and dynamic adjustments that are beyond current AI capabilities; it remains research-stage.
Technical feasibility todayclaude-sonnet-51/5No deployed product trains visually impaired clients in physical mobility skills like cane technique or guide dog handling; this remains an entirely human, in-person professional service.

Train clients to use tactile, auditory, kinesthetic, olfactory, and proprioceptive information.

3

CI 0–5 · exposure 0 · augmentation 25 · importance 4.4/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Vision rehabilitation occurs in low-digitization, human-intensive sectors (healthcare, nonprofit rehabilitation services) with strong regulatory constraints and client vulnerability. Adoption of automation in this sector is extremely slow and limited to administrative tasks, not clinical intervention.
Sector adoption velocityclaude-sonnet-51/5Rehabilitation therapy is a low-digitization, high-touch physical/clinical field with minimal AI agent deployment in production settings.
Augmentation potentialclaude-haiku-4-5-202510012/5AI might assist marginally with assessment documentation, virtual practice environments, or tracking client progress, but the core work—training sensory integration through embodied, adaptive coaching—is not meaningfully augmented by current AI. The therapist's expertise and presence remain irreplaceable.
Augmentation potentialclaude-sonnet-52/5AI could support scheduling, exercise planning, or generating training materials, but offers minimal direct assistance during the actual hands-on sensory training process.
Task automatabilityclaude-haiku-4-5-202510011/5This task requires direct, in-person sensory training and adaptive coaching tailored to individual clients' residual abilities and learning needs. Current AI cannot deliver the embodied, real-time physical guidance, tactile feedback, and dynamic environmental adaptation that define effective orientation and mobility rehabilitation.
Task automatabilityclaude-sonnet-51/5This requires hands-on physical guidance, real-time sensory coaching, and adaptive interpersonal instruction that current AI cannot perform end-to-end.'},
Adoption barriersclaude-haiku-4-5-202510015/5Vision rehabilitation therapists must be licensed professionals, and the task inherently requires direct human contact, real-time responsiveness to client needs, and hands-on guidance that cannot be delegated to automated systems. Legal and professional standards mandate human therapist involvement.
Adoption barriersclaude-sonnet-54/5This work typically requires a credentialed specialist working directly and physically with clients, involving safety, liability, and hands-on skill-building that cannot be delegated to unlicensed or automated means.
Cost vs. human wageclaude-haiku-4-5-202510011/5The loaded cost of a licensed vision rehabilitation therapist vastly exceeds any feasible AI infrastructure cost, and AI cannot yet perform the task at all, making direct cost comparison moot. Any deployed solution would require human therapist oversight and participation.
Cost vs. human wageclaude-sonnet-51/5There is no AI substitute performing this physical training task, so cost comparison favors the human specialist entirely.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed AI product reliably performs sensory rehabilitation training independent of a human therapist. While AI might support assessment or provide some educational content, the core task—training clients to integrate and use alternative sensory modalities—demands skilled human presence and ongoing adjustment to client performance.
Technical feasibility todayclaude-sonnet-51/5No deployed product provides physical training of tactile, kinesthetic, or proprioceptive skills to clients with vision impairment; this remains firmly in-person and manual.

Assess clients' functioning in areas such as vision, orientation and mobility skills, social and emotional issues, cognition, physical abilities, and personal goals.

3

CI 0–5 · exposure 0 · augmentation 25 · importance 4.3/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Vision rehabilitation is a small, specialized clinical field with low digitization and minimal evidence of AI adoption in production settings. The sector is dominated by regulated healthcare providers with institutional caution toward automation of client-facing clinical assessment.
Sector adoption velocityclaude-sonnet-51/5Rehabilitation therapy for vision-impaired clients is a small, highly manual, low-digitization field with minimal AI agent deployment in production settings.
Augmentation potentialclaude-haiku-4-5-202510012/5AI could assist with narrow components such as organizing assessment data or suggesting mobility testing protocols, but the core work—adaptive clinical evaluation, rapport, and synthesis into personalized rehabilitation goals—remains dependent on human therapist expertise and judgment.
Augmentation potentialclaude-sonnet-52/5AI could help with documentation, scheduling, or referencing assessment protocols, but offers limited direct assistance during the hands-on, observational assessment itself.
Task automatabilityclaude-haiku-4-5-202510011/5This task requires comprehensive clinical assessment across multiple domains (vision, mobility, cognition, emotional state, physical abilities, personal goals) that demand nuanced human interaction, clinical judgment, and individualized evaluation. Current AI systems cannot reliably conduct the multi-sensory, adaptive assessments and rapport-building necessary to assess these interrelated areas holistically.
Task automatabilityclaude-sonnet-51/5This requires hands-on clinical assessment involving direct observation of physical mobility, vision function testing with specialized equipment, and interpersonal evaluation of emotional state/goals, none of which current AI can perform end-to-end.
Adoption barriersclaude-haiku-4-5-202510015/5This task involves clinical diagnosis and rehabilitation planning that requires state licensure (occupational therapist or rehabilitation counselor credentials in most jurisdictions) and legal liability for assessment accuracy. Regulatory frameworks and professional licensing explicitly mandate human professional judgment for vision rehabilitation assessment.
Adoption barriersclaude-sonnet-54/5This assessment typically requires a credentialed specialist for legal/clinical documentation, insurance reimbursement, and safety in mobility training, creating strong professional and liability barriers.
Cost vs. human wageclaude-haiku-4-5-202510011/5AI inference and integration costs for vision rehabilitation assessment are not yet viable; the task requires domain-specific clinical expertise, adaptive testing protocols, and personalized evaluation that would demand expensive custom integration, oversight, and validation far exceeding the cost of a trained therapist.
Cost vs. human wageclaude-sonnet-51/5AI cannot substitute for the physical presence and multi-domain clinical judgment required, so there is no viable AI cost comparison—the human specialist remains necessary.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed product performs full vision rehabilitation assessment reliably in production settings. While AI tools exist for limited vision testing (e.g., visual acuity screening), none perform the comprehensive, integrated clinical assessment across mobility, social-emotional, cognitive, and physical domains that this task requires.
Technical feasibility todayclaude-sonnet-51/5No deployed AI product conducts holistic in-person functional assessments combining vision, mobility, cognition, and psychosocial evaluation; this remains squarely a human clinical activity.

Teach clients to travel independently, using a variety of actual or simulated travel situations or exercises.

3

CI 0–5 · exposure 0 · augmentation 25 · importance 4.3/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5This field is small-scale, highly specialized, and involves direct physical service delivery with minimal digitization; adoption of AI in production use is virtually non-existent in vision rehabilitation therapy.
Sector adoption velocityclaude-sonnet-51/5Vision rehabilitation and O&M training occurs in low-digitization, hands-on healthcare/education settings with minimal AI agent deployment or displacement to date.
Augmentation potentialclaude-haiku-4-5-202510012/5AI could assist with pre-training instruction, simulation scenario design, or documenting client progress, but these are peripheral to the core task of teaching physical independence through guided, supervised practice in real or simulated environments.
Augmentation potentialclaude-sonnet-52/5AI-based route-planning apps, GPS navigation aids, and simulation software can support lesson planning or supplement client practice, but they play a marginal assistive role compared to the in-person instruction itself.
Task automatabilityclaude-haiku-4-5-202510011/5Teaching independent travel requires hands-on physical guidance, real-time safety assessment, and dynamic responsiveness to client needs in actual environments. Current AI cannot safely supervise or physically guide a person through travel training scenarios.
Task automatabilityclaude-sonnet-51/5This requires hands-on physical guidance, real-time spatial coaching, and adaptive in-person instruction using canes, guide dogs, or other mobility techniques in actual environments—something current AI cannot perform end-to-end.
Adoption barriersclaude-haiku-4-5-202510015/5Strong legal and professional barriers exist: licensure requirements, legal liability for client safety during mobility training, mandatory human-contact requirements for hands-on guidance, and professional standards requiring credentialed therapists to deliver and sign off on rehabilitation services.
Adoption barriersclaude-sonnet-54/5This is a licensed rehabilitation specialty requiring certified professionals to ensure client safety during real-world travel training, creating strong liability and credentialing barriers to automation.
Cost vs. human wageclaude-haiku-4-5-202510011/5AI integration would be extremely expensive relative to the human therapist cost when accounting for required hardware (robotics or VR), integration, safety systems, and human oversight—far exceeding the loaded wage of a specialist therapist.
Cost vs. human wageclaude-sonnet-51/5There is no AI substitute performing this physical, supervised training task, so the human specialist remains the only viable and thus cost-effective option.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed product reliably performs this task end-to-end. While AI can provide some instructional content or simulation, it cannot replace a therapist's real-time physical presence, safety monitoring, and adaptive coaching during actual or simulated travel.
Technical feasibility todayclaude-sonnet-51/5No deployed product provides embodied orientation and mobility training for visually impaired clients in real or simulated travel environments; this remains a human-delivered therapeutic service.

Teach self-advocacy skills to clients.

3

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

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Vision rehabilitation services are provided by small, specialized, and often non-profit or clinical organizations with low digital maturity and limited AI adoption; the sector is not deploying AI agents for therapeutic or educational intervention roles.
Sector adoption velocityclaude-sonnet-51/5Vision rehabilitation therapy is a small, highly specialized, in-person service sector with minimal AI tool integration or reported adoption in practice.
Augmentation potentialclaude-haiku-4-5-202510013/5AI could assist therapists by drafting educational materials, compiling resources, or organizing client progress notes, raising some productivity on administrative aspects; however, the core work of teaching and mentoring remains human-driven.
Augmentation potentialclaude-sonnet-52/5AI could help generate educational materials or role-play scripts to prepare a therapist's session content, but it plays a minor supporting role rather than transforming the coaching interaction itself.
Task automatabilityclaude-haiku-4-5-202510011/5Teaching self-advocacy skills requires ongoing personalized interaction, reading individual emotional and social cues, and adapting interventions to each client's specific barriers and strengths. Current AI systems cannot reliably replicate the empathetic, adaptive, and relationship-based instruction that is central to this task.
Task automatabilityclaude-sonnet-51/5Teaching self-advocacy requires modeling interpersonal courage, live coaching through emotionally charged scenarios, and adapting to a specific client's psychosocial context, none of which current AI can execute end-to-end.
Adoption barriersclaude-haiku-4-5-202510015/5Vision rehabilitation is a licensed or credential-requiring profession in many jurisdictions, and teaching self-advocacy is fundamentally a human-contact task involving vulnerable clients who rely on trust and personalized guidance; regulatory and legal barriers are high.
Adoption barriersclaude-sonnet-54/5This task typically occurs within a therapeutic relationship often tied to certification/licensure standards and requires trust-building and human judgment, creating strong professional and ethical barriers to automation.
Cost vs. human wageclaude-haiku-4-5-202510011/5A trained therapist's time is essential to this task; the setup, oversight, and liability costs of an AI system attempting to replace or substantially reduce human involvement would exceed the loaded wage of the specialist.
Cost vs. human wageclaude-sonnet-51/5There is no viable AI substitute performing this task, so cost comparison favors the human specialist entirely; any AI attempt would require extensive human oversight negating savings.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed product reliably teaches self-advocacy skills to vision-impaired clients in production. This task requires genuine mentoring, trust-building, and real-world practice feedback that AI has not demonstrated at scale or with acceptable error rates.
Technical feasibility todayclaude-sonnet-51/5No deployed product provides in-person self-advocacy coaching for vision-impaired clients; this remains a human relational skill-building task not addressed by existing AI products.

Teach independent living skills or techniques, such as adaptive eating, medication management, diabetes management, and personal management.

3

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

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Vision rehabilitation is a small, specialized, regulated healthcare sector with strong professional credentialing and low digital transformation. Adoption of AI in this context remains minimal; therapists retain control of client relationships and technique delivery.
Sector adoption velocityclaude-sonnet-51/5Vision rehabilitation and allied health therapy services are a low-digitization, hands-on, small-scale field with minimal AI agent adoption in production settings.
Augmentation potentialclaude-haiku-4-5-202510012/5AI might assist with educational content (e.g., video explanations of techniques) or scheduling, but it cannot augment the core teaching role—demonstration, real-time correction, safety monitoring, and relationship-building. Minimal productivity transformation while the human remains in the loop.
Augmentation potentialclaude-sonnet-52/5AI could help with generating educational materials, reminders, or scheduling, but offers only marginal support for the core hands-on teaching and safety supervision involved.
Task automatabilityclaude-haiku-4-5-202510011/5This task fundamentally requires real-time, personalized instruction adapted to individual disability status, safety considerations, and behavioral change. Current AI systems cannot safely demonstrate techniques, assess physical performance, provide corrective feedback in person, or adapt instruction based on a client's motor control and cognitive capacity in real time.
Task automatabilityclaude-sonnet-51/5This requires in-person, hands-on physical instruction and safety supervision (e.g., using knives, insulin injections, mobility around a stove) with a visually impaired client, which current AI cannot physically perform or supervise.
Adoption barriersclaude-haiku-4-5-202510015/5Vision rehabilitation therapists are typically licensed or credentialed professionals whose training, certification, and liability framework legally tie the task to a qualified human. Additionally, the task involves direct personal contact, medical supervision (diabetes, medication management), and safety assessment—all highly protected by regulation and organizational policy.
Adoption barriersclaude-sonnet-54/5This work often requires certified specialists, involves safety-critical activities (medication, diabetes management) with real injury/liability risk, and depends on physical presence and trust-building with vulnerable clients.
Cost vs. human wageclaude-haiku-4-5-202510011/5AI inference alone cannot substitute for the human therapist's in-person assessment, demonstration, physical guidance, and motivation. The cost of AI oversight and human instruction that remains necessary would exceed the cost of direct human delivery.
Cost vs. human wageclaude-sonnet-51/5AI cannot deliver the physical, supervised instruction required, so there is no viable AI substitute cost to compare against the human specialist's wage.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed AI product reliably teaches adaptive living skills or rehabilitation techniques to vision-impaired individuals. This requires embodied demonstration, hands-on correction, emotional support, and dynamic assessment of motor learning—all outside current AI capability. Chatbots cannot replace the therapist's role in monitoring safety and progress.
Technical feasibility todayclaude-sonnet-51/5No deployed AI product provides hands-on adaptive living skills training for visually impaired individuals; this remains entirely a human, in-person therapeutic service.

Monitor clients' progress to determine whether changes in rehabilitation plans are needed.

3

CI 0–5 · exposure 0 · augmentation 38 · importance 4.1/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Vision rehabilitation is a small, specialized clinical field with limited digitization and slow adoption of information technology. The sector is characterized by individual practices, small teams, and regulatory conservatism, all of which slow AI adoption.
Sector adoption velocityclaude-sonnet-51/5Vision rehabilitation is a small, specialized, hands-on healthcare field with low digitization and minimal AI adoption for clinical progress monitoring.
Augmentation potentialclaude-haiku-4-5-202510012/5AI could potentially assist by tracking and summarizing objective metrics (e.g., mobility speed, task completion times) if collected digitally, but most progress assessment involves qualitative observation, subjective client feedback, and clinical reasoning that resist meaningful augmentation by current systems.
Augmentation potentialclaude-sonnet-53/5AI can help track and analyze progress data, generate documentation, or flag patterns for review, aiding the specialist's decision-making even though it cannot replace the assessment itself.
Task automatabilityclaude-haiku-4-5-202510011/5Monitoring client progress and determining whether rehabilitation plans need adjustment requires deep, individualized understanding of each client's functional capabilities, psychological state, and contextual life circumstances. Current AI systems cannot reliably assess these nuanced human outcomes or make clinical judgments about treatment modifications.
Task automatabilityclaude-sonnet-51/5Monitoring client progress requires in-person observation of physical mobility, adaptive skills, and emotional state, plus clinical judgment to adjust rehabilitation plans, none of which current AI can perform end-to-end.
Adoption barriersclaude-haiku-4-5-202510015/5This task is protected by strong regulatory and professional barriers: vision rehabilitation therapists are licensed specialists, progress monitoring is a core clinical function requiring professional judgment and accountability, and liability exposure is high if algorithmic decisions cause harm or delay needed plan adjustments.
Adoption barriersclaude-sonnet-54/5This is a licensed clinical task requiring professional judgment and often documentation for insurance/regulatory compliance, creating strong barriers to automation.
Cost vs. human wageclaude-haiku-4-5-202510011/5The cost of deploying AI systems to monitor rehabilitation progress, combined with necessary human oversight and clinical validation, would exceed the wage of a vision rehabilitation therapist performing direct assessment, particularly given the low error tolerance for clinical decisions.
Cost vs. human wageclaude-sonnet-51/5There is no viable AI substitute performing this function, so the cost comparison favors the human specialist entirely.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed product reliably performs this task. While general health monitoring AI exists, the specialized assessment of vision rehabilitation progress—requiring observation of orientation, mobility, adaptive behavior, and personal goals—remains firmly in the human clinical domain and is not automated in production systems.
Technical feasibility todayclaude-sonnet-51/5No deployed product autonomously assesses low-vision or mobility rehabilitation progress and revises therapy plans; this remains a hands-on clinical function.

Collaborate with specialists, such as rehabilitation counselors, speech pathologists, and occupational therapists, to provide client solutions.

3

CI 0–5 · exposure 0 · augmentation 38 · importance 3.8/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5Healthcare and rehabilitation services, particularly specialized therapy domains, show slow AI adoption. These fields prioritize human expertise, regulatory compliance, and direct client contact, with minimal evidence of AI displacing collaborative clinical decision-making.
Sector adoption velocityclaude-sonnet-51/5Healthcare/rehabilitation services sectors show slow AI adoption for direct clinical collaboration tasks, especially those requiring multi-disciplinary human judgment.
Augmentation potentialclaude-haiku-4-5-202510012/5AI might assist with scheduling, documentation, or information retrieval to support collaboration, but current systems offer limited meaningful augmentation of the core interpersonal and clinical judgment required to coordinate holistic rehabilitation solutions across multiple specialists.
Augmentation potentialclaude-sonnet-53/5AI tools can assist with scheduling, documentation sharing, and summarizing case notes across specialists, improving coordination efficiency without replacing the collaborative judgment itself.
Task automatabilityclaude-haiku-4-5-202510011/5Collaboration with other specialists inherently requires human judgment, negotiation, and interpersonal coordination. AI cannot autonomously participate in multidisciplinary team meetings or establish therapeutic relationships needed to develop integrated client solutions.
Task automatabilityclaude-sonnet-51/5This is an interpersonal, judgment-based collaborative process among human specialists that requires physical presence, clinical judgment, and real-time coordination that AI cannot perform end-to-end.
Adoption barriersclaude-haiku-4-5-202510015/5Licensure requirements for vision rehabilitation therapists, speech pathologists, occupational therapists, and rehabilitation counselors create hard legal barriers; humans must be licensed to participate in clinical collaboration and treatment planning. Liability and professional accountability are tied to individual practitioners.
Adoption barriersclaude-sonnet-54/5Licensed professionals must be involved in clinical collaboration and decision-making for client care plans, creating strong regulatory and liability barriers to automation.
Cost vs. human wageclaude-haiku-4-5-202510011/5The cost of AI infrastructure, integration, and oversight to facilitate specialist collaboration would exceed the operational savings, particularly given that human specialists must ultimately review and validate any AI-generated coordination or recommendations.
Cost vs. human wageclaude-sonnet-51/5There is no viable AI substitute performing this collaborative clinical task, so cost comparison favors the human entirely.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed AI system can reliably participate as an equal member in professional collaborative teams or independently coordinate complex, personalized rehabilitation solutions across multiple specialist domains in production settings.
Technical feasibility todayclaude-sonnet-51/5No deployed AI product performs interdisciplinary clinical collaboration and case coordination for vision rehabilitation clients; this remains a human professional function.

Teach cane skills, including cane use with a guide, diagonal techniques, and two-point touches.

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-4-5-202510011/5This task occurs in specialized clinical and educational settings (low-vision clinics, rehabilitation agencies) with strong professional licensing, limited digitization, and persistent cultural and regulatory preference for in-person, hands-on training from qualified specialists.
Sector adoption velocityclaude-sonnet-51/5Vision rehabilitation and O&M training is a small, highly specialized, in-person service sector with minimal AI integration or digitization of the core physical training activity.
Augmentation potentialclaude-haiku-4-5-202510012/5AI could assist marginally with video demonstration, progress tracking, or educational materials, but the core skill—teaching safe, effective motor control through physical guidance—remains fundamentally human-dependent and offers limited scope for augmentation.
Augmentation potentialclaude-sonnet-52/5AI could help with scheduling, generating training plans, or tracking client progress notes, but offers little direct assistance during the physical cane-skills instruction itself.
Task automatabilityclaude-haiku-4-5-202510011/5Teaching physical cane skills requires real-time demonstration, hands-on correction of body position and movement, and adaptive feedback based on the learner's proprioceptive and tactile responses. Current AI cannot perform in-person physical instruction or provide the tactile guidance essential to this task.
Task automatabilityclaude-sonnet-51/5This requires hands-on physical demonstration, spotting, and real-time correction of a client's gait and cane movements in physical space, which no current AI system can perform.
Adoption barriersclaude-haiku-4-5-202510015/5Vision rehabilitation therapy is a licensed profession in many jurisdictions, and teaching mobility skills for individuals with vision loss involves direct client contact and legal accountability for safety outcomes. Regulatory frameworks and liability concerns create hard barriers to automation.
Adoption barriersclaude-sonnet-55/5Orientation and mobility instruction typically requires certified specialists and involves direct safety risk to a visually impaired client, creating strong professional and liability barriers to automation.
Cost vs. human wageclaude-haiku-4-5-202510011/5Any AI-assisted approach would require substantial human oversight, correction, and hands-on modeling by a trained therapist, making the combined cost exceed what a human specialist alone would charge for direct instruction.
Cost vs. human wageclaude-sonnet-51/5There is no viable AI substitute delivering this physical, safety-critical instruction, so AI cost comparison is effectively inapplicable and the human remains the only option.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed AI system can reliably teach complex motor skills involving physical positioning, balance, and cane manipulation in real-world settings. While video instruction exists, AI lacks the ability to observe, correct, and adapt to individual learners' physical performance in situ.
Technical feasibility todayclaude-sonnet-51/5No deployed AI product exists that can physically teach mobility cane techniques; this remains entirely a human, hands-on instructional task.

Related occupations — Healthcare Practitioners & Technical

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.