FAccT ACAI audit of 79 U.S. universities: AI governance skewed to rules, weak on feedback and detection due process

Texas A&M University / Cornell Tech / collaborators · Higher education / independent research · United States (79 universities across regions and Carnegie types)

Student Success

Publisher: ACM FAccT 2026
Original publication: 2026-06-25
Source retrieved: 2026-09-21
Event date: 2026-06-25

Read original source

What happened

Human audit of publicly legible AI governance across 79 U.S. universities finds capacity concentrated in policy clarity and faculty support, with feedback mechanisms and detection-tool safeguards underdeveloped; AI research intensity weakly correlates with ACAI governance scores; LLM audits misrank institutions. Why now: newly relevant after the 2026-09-18/19 Student Defense SHAPE coverage that adoption outpaces student protections—this paper quantifies procedural due-process gaps SHAPE warns about.

Why it matters

Student-success AI advising, tutoring, and integrity enforcement all depend on contestable procedures. U.S. public colleges face Fourteenth Amendment and published-procedure risks when detectors drive discipline without human review and recourse.

Evidence and measured results

ACAI-US79 dataset; three Prolific annotators per institution; domains A–D; comparison to CSRankings; LLM replication study showing unstable rankings. Public materials only (lower bound). Link collection Dec 8–30, 2025.

Limitations and uncertainty

Public-legibility lower bound; U.S.-centric; purposive sample; does not measure internal intranet policies or learning outcomes.

Put this evidence to work

Lighthouse Advisory interpretation, grounded in this source. Enriched 2026-09-21; this does not change the original publication date. Each role follows the same checklist so takeaways compare across sources.

Sales

Role takeaway

Customer problem: AI honor codes and detectors expanded faster than contestable due-process infrastructure students and faculty can find and use.
Stakeholders: provost / general counsel
academic integrity office
faculty senate
student government
teaching & learning
Discovery questions: Can a student locate who reviews AI-detection flags and how to appeal within five minutes?
Is there a standing student/faculty AI governance committee?
Are detectors advisory or determinative?
Value hypothesis: Improving public procedural legibility reduces inequitable informal guidance and legal/reputational risk around AI misconduct cases.
Potential engagement: ACAI-aligned governance readiness assessment and remediation roadmap for feedback and detection domains.
Unsupported claims: That high ACAI score equals ethical AI use
That detectors become accurate via policy
Automated LLM audit replacing human review

Pre-sales engineering

Role takeaway

Fit: Policy CMS + integrity case workflow with mandatory human review gates on AI-detection evidence.
Architecture and integration: Link central policy hub, syllabus statement library, committee pages, and case-management system; detectors emit advisory scores only.
Prerequisites: Authoritative content owners
Counsel-approved appeal language
Public vs authenticated content map
Constraints: Public-audit lower bound
U.S. legal framing
Purposive 79-university sample
Security: Misconduct case data needs strict access control; avoid exposing detector vendor models that process student work without DPA.
Proof of value: Before/after ACAI domain scores
Appeal pathway usability test
Zero sole-detector determinations in sample term

Delivery

Role takeaway

Work: Inventory policies, run multi-annotator ACAI checklist, draft detection SOP, launch feedback committee, publish update log.
Dependencies: Counsel
integrity office
web/comms
faculty/student reps
Ownership: Provost-designated AI governance lead.
Skills and adoption: Train instructors that detector scores are investigative leads requiring human judgment.
Governance checkpoints: Public posting of SOP
Committee first meeting minutes
Annual re-audit
Acceptance criteria: Human-review requirement published
Student feedback channel live
Detection tool stance documented
Risks: Shadow AI continues
Resource-thin committees
False confidence from policy text alone

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?

Central public AI governance hub; integrity case systems must treat detector scores as advisory with mandatory human review gates.

Governance

Who approves, reviews and stays accountable for outcomes?

Stand up participatory feedback committees, publish update logs, and specify appeal paths for AI-related academic determinations.

Security and privacy

What data, permissions and controls need testing?

Detection workflows process student writing; govern vendors, false-positive risk for multilingual writers, and access to misconduct case data.

Accessibility and workforce

Who is affected, and what skills or accommodations follow?

Opaque detector governance disproportionately risks multilingual and less-connected students who rely on published rules rather than informal faculty networks.

Procurement

What should contracts, pricing and exit terms secure?

Procurement of detection or student-AI tools should require human-review clauses, appeal documentation, and public policy artifacts—not accuracy marketing alone.

Operating model

Which teams own the service once it runs?

Shift from honor-code updates alone to standing shared-governance structures that revise AI policy with student and faculty input.

Publication history

2026-09-20Student Success · Issue 154 resources

Read preserved resource versions (JSON)

Stable resource ID: teleki-acai-due-process-audit-facct-2026