{"resourceId":"tian-hsi-multilingual-ai-policy-2026","versions":[{"version":"external-2127cf1174f89fa41888a6820a22947a0691b240840dde8a5e9f73fca2f63c80","resource":{"id":"tian-hsi-multilingual-ai-policy-2026","title":"HSI AI policy discourse: English-dominant defaults act as de facto monolingual language policy","organization":"Rutgers University–Newark / University of Texas at San Antonio / Language Policy (Springer)","sector":"Higher education AI governance / language equity at HSIs","geography":"United States (two Hispanic-Serving Institutions)","publishedAt":"Published 17 September 2026","publicationDate":"2026-09-17","eventDate":null,"sourceName":"Language Policy (Springer Open)","sourceLabel":"Tian & Christiansen (2026) — Multilingualism, AI policy, and equity at HSIs","sourceUrl":"https://link.springer.com/content/pdf/10.1007/s10993-026-09803-8.pdf","evidenceClass":"academic-research","outcomeClass":"cautionary","topics":["governance-procurement","accessibility-workforce","operating-model"],"finding":"Critical discourse analysis of AI guidance, syllabus templates, faculty guidelines, task-force materials, and communications at two U.S. HSIs finds: (1) AI governance constructs an implicitly English-dominant user while leaving multilingual practices largely unaddressed; (2) responsible AI use is framed through procedural compliance and individualized responsibility, creating uneven terrain for multilingual users. Authors conclude institutional AI policies function as de facto language policies reproducing monolingual assumptions, and propose a five-dimension continuum for linguistically responsive, participatory AI governance.","sledRelevance":"Published 17 September 2026—same calendar day as the prior edition date and newly available for this 2026-09-18 run. Directly advances student-success equity governance for multilingual HE populations after prior integrity-assistive distinctions.","evidence":"Source findings: CDA patterns across two HSIs; five-dimension continuum proposal. Limitations: two-site discourse study; not an outcomes evaluation of retention/GPA; transferability to non-HSI contexts limited.","architectureImplications":"Interpretation: policy/syllabus systems should encode multilingual guidance and supports rather than English-only defaults.","governanceImplications":"Interpretation: integrate AI governance with institutional language commitments; participatory drafting with multilingual stakeholders.","securityPrivacyImplications":"Policy processes should not require students to paste sensitive work into unvetted consumer tools; provide vetted multilingual-capable institutional options.","caveats":"Discourse analysis ≠ measured enrollment/retention effects. Two HSIs only. Complements but does not replace empirical equity audits.","streamIds":["student-success"],"roles":{"sales":"Sales framing: use this 17 Sept 2026 Language Policy study of two U.S. HSIs to open equity-governance conversations. AI rules constructed an implicitly English-dominant user and framed responsibility as individual compliance, functioning as de facto monolingual language policy. Bound work to participatory policy redesign—not a generic AI code of conduct template. Offer bilingual facilitation for co-design sessions as part of the engagement.","engineering":"Engineering framing: implement policy templates that require language-orientation choices and multilingual resource links. Prerequisites include participatory drafting channels, a corpus of current AI guidance, and bilingual communications capacity. Constraints: evidence is critical discourse analysis at two HSIs, not a causal outcomes study. Proof of value: measurable policy text changes, recorded multilingual stakeholder participation, and syllabus templates that are no longer English-only by default.","delivery":"Delivery framing: delivery is participatory policy work tied to each HSI’s linguistic mission. Ownership sits with equity and academic affairs alongside the AI governance committee. Acceptance requires a five-dimension continuum assessment, revised multilingual syllabus AI templates, and documented co-design participation. Risks include performative checkbox updates that leave English-dominant defaults intact and dumping compliance responsibility onto individual multilingual students."},"roleDetails":{"sales":{"customerProblem":"HSI AI policies often ignore multilingual practice, creating de facto English-only language policy.","stakeholders":["HSI leadership","Student success / equity","Faculty senate","IT/AI governance committee","Language/ESL programs"],"discoveryQuestions":["Do AI policies address multilingual practices explicitly?","Who participated in policy drafting?","Is responsibility individualized without institutional language supports?","How do policies align with HSI linguistic mission?"],"valueHypothesis":"Linguistically responsive AI governance—participatory, multilingual, and integrated with institutional language commitments—reduces monolingual reproduction in student-facing rules.","potentialEngagement":"Policy discourse review at the client HSI against the paper’s five-dimension continuum, with multilingual student/faculty co-design sessions.","unsupportedClaims":["That generic GenAI integrity templates are HSI-ready","That procedural compliance equals equity","That English-default tools are neutral"]},"engineering":{"fit":"Fits AI governance document systems and syllabus-template platforms that must encode multilingual options.","architectureIntegration":"Policy CMS with language-orientation fields; syllabus templates offering multilingual guidance; stakeholder comment workflows; inventory of Spanish/other-language supports.","prerequisites":["Multilingual stakeholder panel","Document corpus of current AI guidance","Alignment check vs HSI language mission statements","Accessible bilingual communications"],"constraints":["CDA of two institutions—not causal outcome study","Discourse findings may not generalize to all HSIs","Technical tooling alone cannot fix participation gaps"],"security":"Policy drafts and comments may include sensitive community input—manage access and avoid feeding into consumer GenAI tools.","proofOfValue":["Multilingual provisions present in policy","Participation record of language stakeholders","Syllabus templates no longer English-only by default"]},"delivery":{"work":"Convene multilingual co-design, revise AI guidance/syllabus templates, and train faculty on linguistically responsive AI use.","dependencies":["Equity/HSI office","Writing/language programs","AI task force","Communications"],"ownership":"HSI academic affairs / equity leadership with AI governance committee.","skillsAdoption":"Faculty and advisors need multilingual AI-guidance skills; committees need discourse-awareness training.","governanceCheckpoints":["Mission-alignment review","Student multilingual advisory input","Publication of revised policy with language continuum"],"acceptanceCriteria":["Five-dimension continuum assessment completed","Revised syllabus AI template multilingual","Documented co-design participation"],"risks":["Checkbox compliance without voice","English-default tools remaining mandatory","Responsibility dumped on individual students"]}},"retrievedAt":"2026-09-19T03:05:14Z","enrichedAt":"2026-09-19T03:11:00Z","enrichmentBasis":"retrieved source","accessibilityWorkforceImplications":"Interpretation: multilingual students and staff need explicit policy recognition and supports; otherwise AI rules shift compliance burdens onto already marginalized language users.","procurementImplications":"Interpretation: evaluate GenAI tools and policy templates for multilingual UX and HSI mission fit—not English-default enterprise chat alone.","operatingModelImplications":"Interpretation: AI governance committees must include language programs and student multilingual representation as standing roles.","sourceVerification":{"responseId":"curl-springer-hsi-lang-policy-09803-8-2026-09-18","openedUrl":"https://link.springer.com/content/pdf/10.1007/s10993-026-09803-8.pdf","referenceExcerpt":"AI governance constructs an implicitly English-dominant user","promptVersion":"sled-research-v4.0","model":null,"basis":"agent-reported inspection"}}}]}