From the Student Success edition of September 15, 2026
Narrative review: HE AI can aid disability access yet risks algorithmic exclusion without UDL governance
Global Journal of Human-Social Science (authors across Brazilian higher-education institutions) · Higher education accessibility / inclusion research · International review with Brazil emphasis (Libras/deaf education); transferability to U.S. SLED requires local legal mapping
- Publisher
- Global Journal of Human-Social Science – G: Linguistics & Education
- Original publication
- Online published 2026-07-18
- Source retrieved
- 2026-09-15
What happened
This peer-reviewed narrative review of purposive 2020–2026 evidence concludes AI can reduce barriers for students with physical, sensory, cognitive, and learning disabilities when paired with accessible design, institutional support, and participatory governance, but systems trained on non-representative data can replicate or amplify inequities across disability, language, race, income, and digital access. Authors call for rights-based UDL, co-design, transparent governance, and continuous evaluation of access, bias, privacy, and learning outcomes—and warn against replacing human Libras interpreters or denying accommodations because a tool is presumed sufficient.
Why it matters
Fills the disability-specific evidence gap repeatedly noted in recent student-success editions. U.S. institutions can transfer the governance checklist with limits: Brazilian Libras cases and narrative (non-meta-analytic) methods are not U.S. causal proof, but the risk pattern for assistive vs exclusionary AI is transferable.
Evidence and measured results
Narrative synthesis (n=20 items in Table 1) spanning peer-reviewed studies, reviews, policy documents, and Brazil-focused Libras/deaf-inclusion work; explicitly not a PRISMA meta-analysis. Online published 2026-07-18 after June 2026 receipt/acceptance.
Limitations and uncertainty
Narrative review without formal risk-of-bias scoring or meta-analysis; heterogeneous evidence; fast-moving tools; authorship includes Brazil-based studies requiring reflexive caution; scarce empirical work on deaf/Libras populations. Not a measured learning or retention trial.
Put this evidence to work
Lighthouse Advisory interpretation, grounded in this source. Enriched 2026-09-15; this does not change the original publication date. Each role follows the same checklist so takeaways compare across sources.
Sales
Role takeaway
- Customer problem
- Institutions adopt GenAI for scale while disability services fear algorithmic exclusion, false accommodation substitutes, and biased detection/support tools.
- Stakeholders
- Disability/accessibility services
- CIO / AI governance
- Academic affairs
- Student disability advocates
- Discovery questions
- Which AI tools are currently treated as accommodation substitutes?
- How are disability and language subgroups tested for bias?
- What co-design processes exist with disabled students?
- Where do human interpreter/mediation requirements remain non-negotiable?
- Value hypothesis
- Accessible AI features may reduce some barriers when governed under UDL and co-design; without those conditions, equity risk can outweigh benefit.
- Potential engagement
- Interpretation: an accessibility-and-AI governance readiness assessment with disability services—not a product deployment sold as inclusion-complete.
- Unsupported claims
- AI replaces interpreters or DSA/ADA accommodations
- Guaranteed equity improvement
- U.S. causal effect sizes from this narrative review
- Bias-free personalization at scale
Pre-sales engineering
Role takeaway
- Fit
- Fit for assistive multimodal features inside governed campus platforms; unfit as opaque high-stakes decisioning about disability accommodations.
- Architecture and integration
- Integrate AI aids with accommodation management systems, human escalation, and audit trails; separate assistive use from integrity surveillance.
- Prerequisites
- Accessibility standards compliance
- Disability-service workflow integration
- Subgroup evaluation data plan
- Consent/transparency for any monitoring
- Constraints
- Narrative evidence limits
- Brazil-to-U.S. legal transfer gaps
- Scarce deaf/Libras empirical studies
- Rapid tool change
- Security
- Treat disability and health-related prompts as sensitive; prohibit silent monitoring; minimize vendor training on accommodation data.
- Proof of value
- Access barrier reduction metrics co-defined with disability services
- Bias tests on disability/language subgroups
- Privacy incident absence
- Human-mediation compliance checks
Delivery
Role takeaway
- Work
- Run co-design workshops, update AI-use and accommodation policies, and stand up continuous access/bias evaluation.
- Dependencies
- Disability services capacity
- Vendor documentation
- Legal/ADA counsel
- Student advisory input
- Ownership
- Interpretation: disability services + AI governance board share ownership; academic units own course-level implementation.
- Skills and adoption
- Train faculty/TAs on assistive AI boundaries and on not denying accommodations because AI exists.
- Governance checkpoints
- Pre-procurement accessibility review
- Co-design sign-off
- Post-deployment bias/access audit
- Annual policy refresh
- Acceptance criteria
- Written non-replacement rules for interpreters/human mediation
- Published evaluation plan for access/bias/privacy/learning
- Student complaint path tested
- Risks
- Tool presumed sufficient to deny accommodations
- Algorithmic exclusion from non-representative data
- Privacy-invasive monitoring
- International evidence over-applied to U.S. compliance
Implementation considerations
Lighthouse Advisory interpretation across the operating dimensions a public-sector buyer must settle before this evidence becomes a design. Each note answers the question under its heading for this specific source.
Architecture and integration
What must connect, and where does the AI sit in the workflow?
Inclusive architectures need multimodal access, human-in-the-loop for high-stakes communication (e.g., interpreters), and bias evaluation on disability/linguistic subgroups. Interpretation: do not treat GenAI captioning or writing aids as complete accommodation substitutes without validation.
Governance
Who approves, reviews and stays accountable for outcomes?
Rights-based UDL, co-design with affected communities, transparent accountability, and continuous evaluation are framed as requirements. Interpretation: accessibility offices and AI governance boards should jointly approve student-facing AI before scale.
Security and privacy
What data, permissions and controls need testing?
Review highlights data sovereignty and monitoring-without-consent risks; AI should not monitor students without transparency. Interpretation: disability-related data are especially sensitive in FERPA/ADA contexts.
Accessibility and workforce
Who is affected, and what skills or accommodations follow?
Core focus: staff must co-design with disabled students, retain human linguistic mediation, and build workforce skills for UDL-aligned AI oversight rather than shifting inclusion labor onto students.
Procurement
What should contracts, pricing and exit terms secure?
Require accessibility evidence, bias testing, data-sovereignty terms, and contractual bars on treating AI as sole accommodation; evaluate vendors on participatory governance support.
Operating model
Which teams own the service once it runs?
Inclusive AI needs institutional responsibility structures—UDL policy, co-design forums, and continuous evaluation—embedded in ordinary student-success operations.
Publication history
- 2026-09-15Student Success · Issue 104 resources
Stable resource ID: gjhss-ai-disability-inclusion-narrative-2026