Higher Education & ResearchIssue 364 records
Student Success
For 2026-10-12, student-success evidence is about AI governance and capacity-building rather than measured outcomes. Eastern Illinois University's Student Senate unanimously asked leadership to regulate faculty AI use, while the president said central policy will cover data privacy only and leave academic rules to departments. A University of Calgary scan of 15 Canadian research universities found only about 9% of 953 AI initiatives were instructor-facing teaching work. Bradley University won a $3 million federal grant pairing advising, early identification and accessibility with first-year AI literacy, and the University of Hartford seeded a guidelines-first responsible-AI hub. Operating takeaway (interpretation): pair AI access with instructor support, a minimal campus-wide disclosure baseline and outcome measurement. Limits: announcements, one student-newspaper report and a non-peer-reviewed scan; no new measured learning or retention results.
- Evidence records
- 4
- Cross-source patterns
- 2
- Evidence classes
- 2 independent reporting1 independent research1 vendor claim
- Outcomes
- 2 cautionary2 emerging
- Source freshness
- 4 new this fortnight
- Research completed
- 2026-10-13
Choose a role to see its takeaway beside every record in the ledger.
Patterns across the evidence
Synthesis · Lighthouse Advisory interpretation
New campus AI investments emphasize literacy, coordination and faculty development, but rarely commit to measured outcomes
In reports dated 12 October, Bradley described a federal grant pairing advising and early identification with first-year AI literacy, an AI Fellows Program and staff development, and Hartford launched a coordination-first AI hub starting with a director, guidelines and faculty development. A same-day scan of 15 Canadian research universities found only about 9% of 953 public AI initiatives were instructor-facing teaching work. Interpretation: institutions are building literacy, coordination and faculty-development scaffolding, but neither announcement states how the AI components will be measured, and instructor-facing work may remain a small share of overall activity. Limits: two announcements and a non-peer-reviewed web scan; no outcome data.
Operating questionWhat share of the institution's AI initiatives and funding directly support instructors and include a measured student outcome?
Supporting evidenceBradley University; U.S. Department of Education Strengthening Institutions ProgramUniversity of HartfordUniversity of Calgary (Werklund School of Education; Hunter Hub) researchers; 15 Canadian research-intensive universities
Where AI rules sit, centrally or in departments, is becoming a student-equity question
At EIU, the Student Senate unanimously asked for a campus-wide baseline on faculty AI use, while the president limited central policy to data privacy and delegated academic rules to departments; students called uneven departmental rules inequitable. The Canadian scan found greater centralization of AI governance did not reliably predict more initiatives or gains. Interpretation: neither full centralization nor full delegation is self-evidently better; a minimal shared baseline such as disclosure plus departmental flexibility may be a workable middle. Limits: one student-newspaper report and a non-peer-reviewed scan.
Operating questionDoes the institution have a minimal campus-wide AI baseline (e.g., faculty disclosure) that students can rely on across departments?
Supporting evidenceEastern Illinois University (Student Senate; Office of the President)University of Calgary (Werklund School of Education; Hunter Hub) researchers; 15 Canadian research-intensive universities
Evidence ledger
Full record · every source keeps its link and limitations
Eastern Illinois University: Student Senate unanimously asks administration to regulate faculty AI use; president's AI policy will cover data privacy only and leave curriculum to departments
The EIU Student Senate unanimously passed a resolution calling on the president and provost to set a campus-wide baseline policy on what faculty can and cannot use AI for, citing student concerns and a petition with 512 online signatures. The same morning, President Gatrell told campus the university's forthcoming internal governing policy on AI would focus solely on data privacy and intentionally not regulate curriculum, deferring to departments and faculty leaders for AI guidance. Student leaders said department-specific rules are inequitable for students and raised academic-integrity concerns; some English assignments have moved back to paper.
Why it matters, evidence and limitations
- Why it matters
- Published 11 October, this is a fresh example of students, rather than faculty, pushing for limits on instructor AI use, and of a campus splitting AI governance into a central data-privacy policy and decentralized academic rules.
- Evidence and measured results
- One student-newspaper report of a Student Senate meeting; quotes from student leaders and the president's campus email. Petition size (512 online signatures) reported; no survey, usage or outcome data.
- Limitations and uncertainty
- Student newspaper; single meeting; resolution is non-binding; the president was not interviewed beyond an emailed statement; policy text unavailable.
Scan of 15 Canadian research universities finds 953 public AI initiatives but only about 9% aimed at instructors' teaching
A systematic scan of publicly visible AI activity at 15 Canadian research-intensive universities between March and June 2026 identified 953 publicly documented AI initiatives across teaching, literacy, credentials, entrepreneurship, partnerships, infrastructure and governance. Only 91, about nine per cent, were explicitly instructor-facing teaching initiatives. Eight of the 15 institutions were monetizing some form of AI education. Institutional size did not reliably predict more initiatives, AI research output or entrepreneurship gains, nor did greater centralization of AI governance. The authors warn institutional responses can become tool-centred: provide a platform, publish guidelines and assume adaptation will follow.
Why it matters, evidence and limitations
- Why it matters
- Published 12 October, it offers a fresh, quantified view that institutional AI activity is concentrated outside instructor support, which bears directly on whether student learning gains from AI are plausible at scale.
- Evidence and measured results
- Web scraping and review of 15 university websites with keyword searches; initiatives categorized into institutional models. Not peer-reviewed; no student, instructor or outcome data.
- Limitations and uncertainty
- Not peer reviewed; counts of publicly visible initiatives may misrepresent capacity; Canadian research universities only; no outcome measures.
Bradley University wins $3 million federal Strengthening Institutions grant for advising, early identification, accessibility and first-year AI literacy
Bradley University has been awarded a multi-year $3 million federal grant through the U.S. Department of Education's Strengthening Institutions Program. Bradley says it will expand academic support, improve advising and accessibility services, help identify students who may need assistance earlier, redesign courses and expand Supplemental Instruction, aiming to improve retention, academic performance and completion. AI literacy will be incorporated into first-year and student success programs, covering responsible use, evaluating AI-generated material, prompting and recognizing misinformation. The grant also funds an AI Fellows Program, faculty and staff professional development and ADA-related training, with plans to sustain programs beyond the grant.
Why it matters, evidence and limitations
- Why it matters
- Published 12 October, it is a fresh federal-funded example of pairing early identification and advising with AI literacy inside first-year student success programming, a pattern also seen in recent Title III awards.
- Evidence and measured results
- University statements relayed by local radio; no baseline retention data, evaluation plan, vendor or tool details.
- Limitations and uncertainty
- Announcement-stage; no outcomes; details secondhand via local media; grant period length not stated.
University of Hartford launches a $100,000-seeded Hub for Human Innovation and Responsible AI, starting with a director to write responsible-use guidelines
The University of Hartford is creating a Hub for Human Innovation and Responsible AI with a $100,000 seed investment from an alumnus and regent. The campus-wide initiative is intended to further student success and faculty innovation and initially serves as a coordinating leadership structure rather than a physical space. Its first step is hiring a director to develop guidelines for responsible AI use. Student programming will focus on AI fluency, experiential learning, innovation and career readiness; faculty and staff will receive professional development and support for integrating AI into teaching, research, student services and operations. The provost framed questions about human judgment, intellectual property, personal information and environmental risk.
Why it matters, evidence and limitations
- Why it matters
- Published 12 October, it is a fresh example of a small private university choosing a coordination-first, guidelines-first approach to AI for student success before deploying tools.
- Evidence and measured results
- University news release only; no metrics, budget detail beyond seed funding or timeline.
- Limitations and uncertainty
- Self-reported announcement; seed funding is modest; no staff hired yet; no outcomes.
How to read this edition
Source findings, measured results and limitations come from the cited publications. Patterns, operating questions, role takeaways and implementation considerations are Lighthouse Advisory interpretation, stated as questions to validate locally rather than guaranteed outcomes. Vendor and operator claims are labeled as claims. Full research method.
- Independent reporting
- Independent reporting with attributable sources but without a formal evaluation design.
- Independent research
- Research conducted outside the implementing organization.
- Vendor claim
- A supplier-provided assertion that has not been upgraded to independent evidence.