AI Governance Policy Review

Prepared: October 2026 · For institutional and public use · Hallmark University, Ijebu-Itele, Ogun State

⇩ Download Full Document (.doc)

Executive Summary

This document constitutes a comprehensive, original policy review commissioned by Hallmark University, Ijebu-Itele, Ogun State, as a foundational step in the development of the university’s Artificial Intelligence (AI) Governance Policy. The review examines two distinct but instructive case studies – Afe Babalola University, Ado-Ekiti (ABUAD), representing the evolving landscape of AI governance within Nigeria’s private university sector, and Yale University, USA, representing a globally benchmarked model of sophisticated, well-resourced, and ethically grounded AI governance in higher education.

The review finds that both institutions, despite vastly different scales of operation, converge on a central truth: AI governance in a university is not merely a technical exercise – it is a moral, pedagogical, and institutional statement of identity. Every policy decision about AI use reflects what a university believes about learning, honesty, human dignity, and its responsibility to the society it serves.

Drawing from both case studies, this document proposes a contextually appropriate, forward-looking AI Policy Framework that positions Hallmark University as a responsible, innovation-embracing institution – one that meets and anticipates NUC regulatory expectations, protects academic integrity, empowers its faculty and students, and contributes meaningfully to the national conversation on AI and education in Nigeria.

Table of Contents

1. Introduction: The Mandate for an AI Policy at Hallmark University
2. The Global Landscape of AI Governance in Universities
3. Deep Review: Afe Babalola University (ABUAD) AI Policy
4. Deep Review: Yale University AI Policy Ecosystem
5. Comparative Analysis: ABUAD vs. Yale
6. Implications and Lessons for Hallmark University
7. Proposed AI Policy Framework for Hallmark University
8. Conclusion
9. References and Related Documents

1. Introduction: The Mandate for an AI Policy at Hallmark University

Hallmark University, Ijebu-Itele, is a private university in Ogun State, Nigeria, committed to producing graduates who are not only technically competent but morally grounded and globally competitive. In an era increasingly shaped by Artificial Intelligence, this commitment demands that the university develop a deliberate, principled, and practical stance on how AI technologies are adopted, governed, and taught within its walls.

The question facing Hallmark University is no longer whether to engage with AI – that decision has already been made by history. ChatGPT alone crossed 100 million users within two months of its launch in November 2022, making it the fastest-adopted technology in human history. Today, students across Nigerian universities – including Hallmark University – are already using AI tools for everything from assignment drafting to research summaries to examination preparation, with or without institutional guidance. The real question, therefore, is whether Hallmark University will shape how its community engages with AI, or be shaped by it.

An institutional AI policy answers that question decisively. It is a document that declares: we have thought carefully about this technology, we understand its promise and its dangers, and we have chosen – as a community – how we will use it with integrity.

1.1 Why Now? The Urgency of Policy Timing

The urgency of this task is compounded by three converging realities. First, the National Universities Commission (NUC) is actively developing its own regulatory framework for AI in Nigerian universities. The NUC’s Towards Responsible AI Integration in the Nigerian University System and its Roadmap for the Integration of Artificial Intelligence into the Nigerian University System – both products of the landmark July 2024 National Policy Dialogue in Abuja – signal that AI governance will increasingly become a criterion in programme accreditation and institutional quality assessments. Universities that arrive at the NUC’s table with a well-developed, publicly documented AI policy will be far better positioned than those scrambling to comply after the fact.

Second, academic integrity has become a genuine crisis in Nigerian higher education in the AI era. Independent research has documented that large proportions of Nigerian university students are using generative AI not as a learning tool but as a submission tool – generating full assignments, paraphrasing answers to evade plagiarism detection, and producing work that does not reflect their own thinking or competence. Without a clear institutional policy, Hallmark University’s lecturers are left to make individual, inconsistent, often uninformed decisions about what constitutes academic misconduct. That inconsistency is unfair to students, dangerous to institutional credibility, and unsustainable.

Third, the competitive landscape of Nigerian higher education is shifting rapidly in favour of institutions that can demonstrate digital preparedness and innovation leadership. ABUAD’s Senate approval of an AI policy in 2026 positioned it as a pioneer. Hallmark University, by developing a thoughtful, published AI governance framework, joins and advances that forward-looking cohort.

1.2 Purpose and Scope of This Review

This review was designed with one primary purpose: to learn comprehensively from two well-studied cases – ABUAD and Yale – so that Hallmark University does not build its AI policy from scratch on untested assumptions, but rather from a position of informed, comparative insight. The review does not seek to transplant either institution’s policy wholesale into Hallmark University’s context. Nigerian private universities face a particular configuration of resource constraints, regulatory obligations, student demographics, and infrastructural realities that neither ABUAD’s nor Yale’s policy fully addresses. What this review offers is a rich source of principles, structures, mechanisms, and cautionary insights that Hallmark University’s policy team can adapt, combine, and localize into an AI governance framework that is authentically its own.

2. The Global Landscape of AI Governance in Universities

2.1 The Scale and Speed of Change

Artificial Intelligence, and specifically generative AI, has introduced a rate of technological change that outpaces the usual cycles of institutional adaptation in universities. Most universities develop policy in response to demonstrated need – and with AI, the need materialized almost overnight. What is striking about the global university response to AI is not simply its volume but its variety: institutions have produced governance frameworks ranging from single-paragraph faculty guidance notes to hundred-page strategic reports, from blanket bans on AI in coursework to mandatory AI integration across curricula. This diversity reflects genuine uncertainty about what responsible AI governance looks like in practice, and it underscores the value of learning carefully from early adopters before making one’s own decisions.

2.2 Key Dimensions of University AI Governance

Across institutions that have developed substantive AI policies, a common set of governance dimensions has emerged. These dimensions form the analytical lens through which this review examines both ABUAD and Yale:

• Ethics – The values and principles that define responsible AI use: fairness, transparency, human dignity, non-maleficence.
• Academic Integrity – How AI use intersects with standards of honest scholarship, authorship, and assessment.
• Data Governance – Protection of institutional, student, and research data in the context of AI platform interactions.
• Faculty Roles – The responsibilities of lecturers in communicating, enforcing, and modelling AI use norms.
• Curriculum – How AI knowledge and skills are embedded into academic programmes and graduate outcomes.
• Enforcement – The mechanisms through which policy violations are identified, adjudicated, and penalized.
• Research Integrity – AI’s intersection with research methodology, publication ethics, and data analysis.
• Equity – Ensuring that AI adoption does not deepen existing divides between students of different socioeconomic backgrounds.
• Governance Structure – The institutional bodies responsible for AI policy oversight, review, and implementation.

2.3 The African and Nigerian Context

The challenge of AI governance in African universities cannot be understood without acknowledging the specific infrastructural, regulatory, and cultural conditions that distinguish this context from global norms. Internet penetration remains uneven across Nigerian university campuses. Power supply instability limits sustained engagement with cloud-based AI platforms. Many students access AI tools via mobile devices rather than computers, affecting the quality and depth of their engagement. Funding disparities between public and private universities create inequities in access to premium AI tools that come with security and privacy assurances that free tools do not offer.

At the same time, African universities face a unique dimension of risk from AI that is rarely discussed in Western policy literature: the risk of intellectual and cultural dependency. When the large language models that power most generative AI tools have been trained predominantly on English-language, Western-authored texts, their outputs inevitably reflect the assumptions, framings, and values of those traditions. For Nigerian students writing dissertations on public administration, social policy, indigenous governance, or development economics, uncritical reliance on AI-generated content risks producing scholarship that is linguistically fluent but culturally shallow – a homogenization of knowledge that contradicts the very purpose of African higher education.

Any AI policy designed for a Nigerian university must grapple honestly with this challenge. It must do more than regulate how students submit assignments. It must affirm what kind of thinkers the university is committed to producing.

3. Deep Review: Afe Babalola University (ABUAD) AI Policy

3.1 Institutional Context

Afe Babalola University, Ado-Ekiti (ABUAD), was established in 2009 by legal luminary Aare Afe Emmanuel Babalola, OFR, CON, SAN, and has grown in under two decades into Nigeria’s most internationally recognized private research university. In the 2026 Times Higher Education (THE) Impact Rankings, ABUAD was ranked 72nd globally, 2nd in Africa, and 1st in Nigeria – a position it has held for five consecutive years. The university operates multiple colleges spanning law, sciences, engineering, social and management sciences, medicine, pharmacy, and environmental sciences, with a growing portfolio of postgraduate programmes.

ABUAD’s institutional character is defined by its founder’s deep commitment to excellence, discipline, and self-sufficiency. The university was built on a single campus in Ado-Ekiti with the deliberate intention of creating a fully residential, tightly governed academic community – one where standards are not aspirational but operational. This institutional DNA is highly relevant to understanding the character of its AI Policy: when ABUAD adopts a framework, it does so with the institutional will to enforce it across a cohesive community.

3.2 The Policy Genesis: A Milestone for Nigerian Higher Education

ABUAD’s AI Policy was approved by the University Senate in June/July 2026, making it – to the best of current evidence – the first formally Senate-approved AI governance policy at a private Nigerian university. The significance of this achievement cannot be overstated. The Senate is the supreme academic authority of any Nigerian university, and Senate approval of the AI Policy elevates it to the same normative standing as the university’s academic regulations, its codes of student conduct, and its research ethics framework. This is not a departmental guideline or a Provost’s advisory note – it is an institutional law with the full weight of the university’s governance structure behind it.

The announcement was made publicly by the Vice-Chancellor, Prof. E. Smaranda Olarinde, FCArb, FCAI, on 7 July 2026, during a visit by the Ekiti State Commissioner for Innovation, Science and Digital Economy:

“Just one week ago, the University’s Senate approved ABUAD’s Artificial Intelligence Policy. The policy provides a comprehensive ethical framework for the responsible integration of Artificial Intelligence across our academic and administrative ecosystem. This is another major step in our journey towards becoming a fully digital and globally competitive Smart University.” — Prof. E. Smaranda Olarinde, Vice-Chancellor, ABUAD, July 7, 2026

The framing of the policy as an “ethical framework” – rather than a technical rulebook or a procedural manual – is a deliberate and important choice. It signals that ABUAD is not approaching AI governance as a compliance exercise, but as a value-driven declaration of institutional purpose. This framing is consistent with the best international practice in university AI governance, and it sets a tone that Hallmark University should seriously consider emulating.

3.3 The Five-Domain Architecture of ABUAD’s AI Policy

Based on the Vice-Chancellor’s announcement and the broader institutional context, ABUAD’s AI Policy is structured around five operational domains. Each domain reflects a discrete area of university activity in which AI’s presence creates both opportunity and risk:

• Domain 1 – Teaching and Learning: Governs how faculty design and deliver instruction with AI, and how students may use AI in their academic work. ABUAD’s ethical-framework stance creates space for course-level differentiation rather than a binary permit-or-prohibit rule.
• Domain 2 – Research: Addresses citation hallucination, methodological opacity, intellectual property confusion, and publication ethics violations by embedding AI governance within existing research-ethics architecture.
• Domain 3 – Administration: Extends the ethical framework to student records, admissions, HR, and financial forecasting – governing institutional data, employee privacy, and decisions affecting real people.
• Domain 4 – Innovation: Affirms AI as a frontier to be explored, not merely managed – positioning students and faculty as producers of AI-powered solutions.
• Domain 5 – Data Governance (Implicit): Read in conjunction with Nigeria’s Data Protection Act (2023) and the NDPR (2019).

3.4 ABUAD’s Policy in the Nigerian Context: Precedence and Responsibility

What makes ABUAD’s AI Policy particularly instructive for Hallmark University is not simply its content but its timing and its significance within the Nigerian higher education ecosystem. Research published as recently as 2024/2025 confirmed that, at that point, no Nigerian or Ghanaian university had a formally published, implemented institutional AI policy. ABUAD’s Senate approval in 2026 changes that picture for the private university sector, and it creates both a benchmark and a degree of competitive pressure for peer institutions.

For Hallmark University, this means two things simultaneously. On one hand, there is a Nigerian-specific model to learn from – one developed in conditions more similar to Hallmark’s own than anything Yale or Harvard has produced. On the other hand, Hallmark University’s AI policy, when developed, must not merely replicate ABUAD’s framework: it must speak with its own institutional voice, reflect its own values, and address the specific characteristics of its student community, its academic programmes, and its geographic context in Ogun State.

3.5 Key Lessons from ABUAD for Hallmark University

1. Senate Authority Matters: Anchoring the AI Policy in Senate approval gives it the highest possible institutional standing. Hallmark University should pursue Senate endorsement rather than settling for a management-level directive.
2. Ethics as the Foundation: Framing the policy as an “ethical framework” rather than a list of rules creates a more durable, adaptable, value-consistent governance instrument.
3. Scope Breadth is Essential: Covering teaching, research, administration, and innovation in a single policy avoids governance gaps.
4. Policy Visibility is Non-Negotiable: Hallmark University must publish its AI policy in full, in plain language, on its website and in student and staff handbooks, from Day One.

4. Deep Review: Yale University Artificial Intelligence Policy Ecosystem

4.1 Institutional Context

Yale University is one of the world’s oldest and most academically distinguished research universities, founded in 1701 in New Haven, Connecticut. With an endowment exceeding $40 billion, a faculty of over 4,000, and 14 schools spanning every major field of human knowledge, Yale operates at a scale and with a resource base that is frankly not comparable to any Nigerian university today. It would therefore be a serious analytical error to treat Yale’s AI policy as a direct template for Hallmark University.

What makes Yale valuable as a case study is not its scale or its budget – it is its intellectual approach to AI governance. Yale does not simply ask: “How do we prevent students from cheating with AI?” It asks: “What does it mean to educate human beings in an age where AI exists – and how must a university as a social institution evolve to fulfil its purpose in that age?” That question is one that Hallmark University, regardless of its resources, must also answer.

4.2 The Architecture of Yale’s AI Governance: A Layered System

Yale’s AI governance operates not as a single document but as an ecosystem of interlocking policies, structures, tools, and practices that reinforce one another:

• December 2022: ChatGPT launches publicly; Yale’s Poorvu Center for Teaching and Learning begins fielding faculty enquiries, with a guidance website live by January 2023.
• August 2023: The Provost issues formal guidance letters to all faculty ahead of the new semester.
• September 20, 2023: Provost-level “Guidelines for the Use of Generative AI Tools” – Yale’s first formal institutional policy document – distributed university-wide.
• January 2024: Yale Task Force on AI (YTAI) established by Provost Scott Strobel; 18 senior faculty and leaders appointed; school-level faculty AI panels convened across all 21 Yale schools.
• June 2024: Task Force Report published – a 115-page document synthesizing input from over 100 faculty, staff, and student consultations.
• August 2024: $150 million strategic commitment announced for AI infrastructure, tool access, faculty development, curricular innovation, and research capacity over five years; the Clarity platform deployed for all Yale community members.
• 2025–2026: Ongoing implementation – AI courses for non-STEM students, a campus-wide AI research symposium, an AI Review Framework, and expanded Library AI Literacy programming.

4.3 The Yale Task Force: A Model of Consultative Policy Development

The Yale Task Force on AI models something rare in Nigerian university governance: the idea that a policy governing the entire university community should be built through genuine, extensive consultation before it is finalized. The Task Force spent six months in active conversation with over a hundred people across every school, discipline, and administrative domain, asking not “what are you doing with AI?” but “what should Yale be doing with AI, given its unique position, its values, and its responsibility to society?”

The 18 Task Force members were deliberately chosen to represent the breadth of Yale’s intellectual community – a philosopher, a neurosurgeon, an astronomer, an expert in African American studies, a computer scientist, a law professor, a librarian, and a public health specialist, among others – sending a clear message: AI is not a technology problem to be solved by the IT department, but a civilizational challenge requiring every dimension of human knowledge.

4.4 The Provost Guidelines: A Baseline of Shared Obligations

Yale’s foundational policy document establishes orienting principles rather than an exhaustive list of permitted and prohibited activities:

• Protect Yale’s confidential information and your own – treat everything shared with an AI tool as if it will become public.
• Always follow academic integrity guidelines and institutional standards of conduct – AI does not create a separate moral universe.
• Faculty must provide clear instructions on permitted AI use and attribution requirements for each course.
• Students must follow their instructor’s AI guidelines for coursework – where no explicit guidance exists, students should assume AI is not permitted.
• Always review and verify AI outputs before use, especially before publication – the human author remains responsible for the accuracy of work submitted under their name.

The default position – that where no explicit authorization exists, AI use for assessed work is not permitted – is a critical design choice, reflecting a sophisticated understanding of how ambiguity operates in real academic communities.

4.5 Academic Integrity: Yale’s Nuanced, Non-Binary Approach

Yale’s policy does not say “AI is permitted” or “AI is prohibited.” It delegates that judgment to individual faculty within a structured framework. The Poorvu Center’s three-tier framework – developed with Cornell University – offers a practical vocabulary:

• Tier 1 – Prohibit: No AI use permitted. Appropriate when constructing an answer unaided is itself the learning objective.
• Tier 2 – Permit with Documentation: AI may be used, but must be disclosed, described, and attributed, exactly as one would acknowledge a human research assistant.
• Tier 3 – Integrate: AI use is designed into the learning experience, with students expected to interrogate, critique, extend, or build upon AI-generated outputs.

Yale College Regulations require students to cite AI as a source when AI-generated language, images, or ideas appear in submitted work. Yale has documented cases in which students were suspended and had grades reduced for AI misuse – policy without precedent is merely aspiration.

4.6 Data Governance: The Silent Crisis in AI Policy

Every time a Yale student, faculty member, or staff member types a query into a public AI tool, they potentially transfer information to a third-party server governed by commercial terms of service that may not align with academic or legal privacy standards. Yale addresses this through a formal risk-classification system distinguishing public, internal, sensitive, and restricted data, and prohibits sensitive or restricted institutional information from entering public AI platforms without appropriate data-processing agreements. The Clarity platform was engineered specifically for high-risk data. For Hallmark University, the lesson is not “build your own AI platform” but “develop a data classification framework that tells your community clearly what they should never put into a public AI tool.”

4.7 Research Integrity and the AI–Publication Interface

Major journals have adopted policies ranging from mandatory AI disclosure to outright prohibition of AI authorship. A researcher who submits a paper without disclosing AI use, where the journal prohibits it, is not merely violating university policy but potentially committing research misconduct. Yale empowers researchers with information rather than imposing a single blanket rule, since appropriate standards of disclosure and attribution differ across disciplines.

4.8 The Philosophical Spine: AI for Humanity

Yale asserts that universities have a distinctive and irreplaceable role in the development of AI – not as technology consumers, but as independent, critical voices that can ask the questions the market cannot. The Yale Task Force’s sixth core recommendation, “AI for Humanity,” captures this: the questions AI raises about consciousness, creativity, authorship, labour, democracy, inequality, and what it means to be human are precisely the questions universities have spent centuries developing the intellectual tools to address.

4.9 Key Lessons from Yale for Hallmark University

1. Policy Must Be a System, Not a Document: Effective AI governance requires a coordinated ecosystem of guidelines, structures, tools, training resources, and enforcement mechanisms.
2. Consult Broadly Before Deciding: Yale’s consultative process produced a policy with genuine community ownership.
3. The Default Matters: “AI is not permitted unless explicitly authorized” protects academic integrity without a blanket prohibition.
4. Data Governance Deserves Its Own Section: Clear, specific language on what must never be entered into a public AI platform.
5. Invest in AI Literacy: Policy without capacity-building is futile.
6. Claim the Philosophical High Ground: Position the university as shaping AI’s impact on society, not merely managing AI risk.

5. Comparative Analysis: ABUAD and Yale

5.1 The Comparative Matrix

5.2 Reading the Comparison Honestly

A comparison between ABUAD and Yale risks producing two misleading conclusions: either that ABUAD’s policy is inadequate because it does not match Yale’s depth, or that Yale’s model is irrelevant to Nigeria because of the resource gap. Both conclusions would be wrong.

ABUAD’s AI Policy is not inadequate – it is young. Senate approval of a comprehensive ethical framework in 2026, while no other Nigerian private university had formally done so, is a genuine achievement reflecting institutional seriousness and governance maturity. The gaps in ABUAD’s public policy communication are almost certainly gaps in communication rather than in institutional thinking.

Equally, Yale’s model is not irrelevant to Nigeria – but it must be read as a source of principles and design patterns, not a blueprint. The right way to use this comparison is to extract the underlying principles that both institutions affirm – ethics as foundation, consultation before finalization, broad scope, faculty responsibility, mandatory attribution, enforcement with consequence, equity as a design principle – and then ask, for each principle: how would this look, and how would this work, at Hallmark University, in 2026?

6. Implications and Lessons for Hallmark University

6.1 What Hallmark University Must Decide

Before any policy language is drafted, Hallmark University’s leadership, faculty, and students need to reach clarity on a set of foundational questions whose answers will give the eventual AI policy its institutional character:

• What is Hallmark University’s position on AI as a learning tool – categorically different from a dictionary, calculator, library, or tutor, or a matter of degree and context to be managed?
• What kind of graduates is Hallmark University committed to producing, and does the policy create conditions in which students actually develop those capacities?
• How does Hallmark University understand the relationship between AI and African knowledge – a neutral tool, or a technology carrying embedded assumptions and cultural biases?
• What level of enforcement is Hallmark University prepared to sustain? A policy that cannot be enforced is worse than no policy.

6.2 Contextual Factors Specific to Hallmark University

• Student Demographics and Digital Equity: Varying socioeconomic backgrounds, digital literacy, and device quality mean any policy presupposing equal access to premium AI tools will create a two-tier academic environment unless institutional access or fair assessment design compensates for it.
• Faculty Capacity and Training: Before faculty can enforce an AI policy, they need to understand AI themselves – this must be accompanied by a comprehensive faculty development programme.
• Infrastructure Constraints: Hallmark University cannot replicate Yale’s GPU clusters, but it can negotiate institutional licensing, leverage funding such as the TETFund Digital Innovation Support Fund or Google.org Africa programmes, and design policy that maximizes whatever tools are accessible.

7. Proposed AI Policy Framework for Hallmark University

Based on the foregoing analysis, this section proposes a comprehensive AI Policy Framework for Hallmark University, Ijebu-Itele. This framework is original – it does not reproduce either ABUAD’s or Yale’s policy, but draws on the principles extracted from both case studies and adapts them to Hallmark University’s specific institutional identity, context, and aspirations.

7.1 Foundational Philosophy: The Hallmark University Position on AI

Hallmark University affirms that Artificial Intelligence is a transformative technological development that will shape every dimension of professional and civic life for the graduates it produces. The university’s response to AI is therefore not primarily defensive – it is educational. At the same time, AI does not alter the fundamental purpose of a university education: to develop in each student the capacity for independent thought, rigorous analysis, clear expression, ethical judgment, and informed action. Where AI removes the productive struggle through which genuine learning occurs, it becomes an obstacle to education rather than an instrument of it.

The Hallmark University AI Policy is therefore built on three pillars:

• Integrity – Every member of the Hallmark community is responsible for the honesty and authenticity of the work they submit under their name.
• Competence – Hallmark University’s graduates must be genuinely capable, without depending on technological assistance as a crutch.
• Wisdom – Knowing how to use a powerful tool responsibly requires understanding its nature, limitations, risks, and consequences for others.

7.2 Governance Structure

7.2.1 Approval Authority: The Hallmark University AI Policy shall be approved by the University Senate and reviewed by the University Council, following the precedent set by ABUAD.

7.2.2 AI Governance Committee: A standing Hallmark University AI Governance Committee (HU-AGOC) shall be established, reporting directly to the Deputy Vice-Chancellor (Academic), comprising:

• The Deputy Vice-Chancellor (Academic) – Chair
• The Registrar or a designated representative
• Two senior faculty members, one from the sciences and one from the social sciences
• The University Librarian
• A student representative
• The Director of ICT
• A legal adviser with expertise in data protection law

HU-AGOC shall meet at minimum once per semester, review policy implementation, address emerging issues, present an annual AI Governance Report to Senate, and serve as the first-instance review body for novel or contested AI-related academic integrity cases referred by Faculty Boards.

7.3 Scope of the Policy

The Hallmark University AI Policy applies to all members of the university community – undergraduate students, postgraduate students, academic staff, research staff, and administrative staff – in all activities connected to the university, covering:

• Academic coursework and assessment
• Research activities, including theses, dissertations, and faculty-led research
• Administrative functions and records management
• Institutional data management
• Teaching and curriculum design

7.4 Academic Integrity Provisions

7.4.1 The Default Position: In the absence of explicit written authorization from a course lecturer, the following constitute a violation of Hallmark University’s academic integrity standards:

• Using any AI tool to generate, in whole or substantial part, work submitted as the student’s own for academic credit
• Using AI to paraphrase or reformulate work that is not the student’s own, to evade plagiarism detection
• Using AI during any examination or timed assessment without explicit written permission
• Submitting AI-generated content without disclosure and attribution, even where AI use is generally permitted

7.4.2 The Three-Tier Instructional Framework: Every course lecturer must communicate their AI policy in writing at the start of each semester, locating the course in one of three tiers:

• Tier A – No AI Use Permitted: The course requires unaided competence (e.g. first-year composition, foundational mathematics, clinical examination skills).
• Tier B – Permitted with Mandatory Disclosure: AI-assisted elements must be disclosed in an AI Contribution Statement; the student remains responsible for accuracy and originality.
• Tier C – Integrated AI Use: AI tools are actively incorporated into the learning design.

AI Contribution Statements under Tiers B and C must specify which tools were used, for what purpose, how outputs were evaluated or modified, and what the student contributed beyond the AI’s contribution.

7.4.3 Enforcement and Penalties:

• Level 1 – Inadvertent Non-Disclosure: Formal written warning, mandatory AI literacy training, discretionary mark reduction.
• Level 2 – Deliberate Misattribution: Zero for the assessment, redo under supervised conditions, one-year academic-record notation.
• Level 3 – Systematic AI Fraud: Referral to the University Disciplinary Committee; potential suspension or dismissal; permanent transcript notation.

7.5 Data Governance and Privacy

The following categories of information must never be entered into any public AI platform (ChatGPT, Gemini, Claude, Copilot, or equivalent) without explicit written authorization from HU-AGOC:

• Student academic records, grades, disciplinary history, or personal identifying information
• Unpublished research data, particularly involving human participants
• Confidential institutional financial, legal, or strategic information
• Staff employment records, performance data, or medical information
• Examination questions, marking guides, or assessment security materials
• Any information whose disclosure would violate the Nigeria Data Protection Act (2023), the NDPR (2019), or any confidentiality obligation

Hallmark University shall develop and publish an Institutional Data Classification Framework within 12 months of this policy’s adoption.

7.6 Faculty Responsibilities and Development

Academic staff must complete the Hallmark University AI Literacy Training Programme within six months of the policy’s commencement; include a clear AI policy statement in every course outline; design assessments that are robust in the context of AI availability; stay informed about AI developments and publication policies in their discipline; and refer suspected violations to their Faculty Board. The university commits to supporting these responsibilities through orientation, workshops via the Centre for Teaching, Learning and Innovation, access to an approved AI tool, and an advisory channel through HU-AGOC.

7.7 Student Rights and Responsibilities

Students have the right to clear, written information about the AI policy for every course, the right to appeal academic integrity decisions through the established appeals process, and the right to access AI literacy training. They are responsible for understanding each course’s policy, for the accuracy and integrity of their submitted work, for disclosing AI use as required, and for developing the genuine intellectual capacities their degree is intended to certify.

7.8 AI Literacy Programme

• Foundation Level (Year 1, all students): What AI is and how it works; Hallmark University’s AI policy; responsible use; data privacy basics; detecting AI limitations and errors. Mandatory 2-hour orientation module.
• Intermediate Level (Year 2 and above): Discipline-specific applications and ethics; academic citation of AI; research integrity; evaluating AI output quality; AI and employment futures. Delivered through faculty-integrated coursework.
• Advanced Level (Postgraduate students and faculty): AI in research design and methodology; publication ethics; AI and IP law in Nigeria; AI governance leadership. Delivered through workshops and research seminars.

7.9 Research Integrity Provisions

Postgraduate students and research staff using AI in any aspect of their research must disclose this in their methodology section, specifying the tool, purpose, and evaluation process. Supervisors are responsible for ensuring compliance. Where a target journal or conference has a specific AI disclosure or prohibition policy, that policy takes precedence over Hallmark University’s general permission structure.

7.10 Equity, Inclusion and Accessibility

Hallmark University commits to prioritizing institution-wide access to at least one approved, data-protected AI tool within the first two years of policy implementation; ensuring no assessment penalizes students for lacking access to tools the university has not made universally available; monitoring the impact of AI adoption on equity across student cohorts, reporting findings annually to Senate; and ensuring AI literacy training is accessible to students with disabilities and limited digital experience.

7.11 Policy Review and Evolution

The AI Policy shall be reviewed formally by HU-AGOC every 12 months, with a full Senate review every two years. Interim guidance may be issued by HU-AGOC between scheduled reviews when significant developments make it necessary.

7.12 Summary of Priority Recommendations

1. Pursue Senate Approval Immediately – submit the Framework to Senate within the current academic session, following ABUAD’s precedent.
2. Establish HU-AGOC – constitute the committee immediately upon approval, with clear terms of reference and a mandate to report annually to Senate.
3. Publish the Full Policy Text – on the website, in the Student and Staff Handbooks, and in every Course Outline from Day 1.
4. Launch the AI Literacy Programme – deliver the Foundation Level module as mandatory orientation from the first semester of adoption.
5. Procure an Institutional AI Tool – negotiate institutional access to a data-protected AI platform (e.g. Microsoft Copilot for Education or Google Gemini for Education).
6. Develop the Data Classification Framework – within 12 months, aligned with the Nigeria Data Protection Act (2023) and NUC guidelines.
7. Engage with the NUC Proactively – submit this review and the adopted Framework to the NUC as a formal institutional communication.
8. Integrate AI into Curriculum Development – incorporate AI competencies into at least one core course per programme by the 2027/2028 session.

8. Conclusion

The development of an AI governance policy is one of the most consequential institutional decisions Hallmark University, Ijebu-Itele, will make in this decade. It is consequential not primarily because of what it regulates, but because of what it says – about who Hallmark University is, what it stands for, and what kind of future it is working to create for its students and for Nigeria.

ABUAD demonstrates that a Nigerian private university can and should take AI governance seriously at the highest institutional level, framing it as an ethical project rather than a compliance exercise – and that adopting a policy is only the beginning; the harder work is operationalizing it. Yale demonstrates what AI governance looks like when built over time, with genuine resources, through authentic consultation, and with clear intellectual ambition – and that “how do we govern AI?” is inseparable from “what is a university for?”

Hallmark University’s AI Policy, when adopted, will not be ABUAD’s policy. It will not be Yale’s policy. It will be Hallmark University’s own – shaped by the specific character of this institution in Ijebu-Itele, Ogun State; by the particular students who walk through its gates; by the specific disciplines it teaches and research its faculty pursue; and by the particular vision of human formation and social contribution that defines what a Hallmark education means.

That policy, thoughtfully developed and faithfully implemented, will be one of the most important contributions Hallmark University can make to the future of Nigerian higher education. This review is offered as a contribution to that work – not as a final word, but as a well-researched starting point from which Hallmark University can build something genuinely its own.

9. References and Related Documents

A. Institutional Primary Sources

• Afe Babalola University, Ado-Ekiti (ABUAD). (2026, July 7). ABUAD on Course of Leading as Globally Renowned Smart University, Says VC. Office of the Vice-Chancellor, ABUAD.
• ABUAD College of Postgraduate Studies. (2025, September). Policy Handbook of the College of Postgraduate Studies. Afe Babalola University, Ado-Ekiti.
• Yale University, Office of the Provost. (2023, September 20). Guidelines for the Use of Generative AI Tools.
• Yale University, Office of the Provost. (2023, August 14). Updated AI Guidance from the Poorvu Center.
• Yale Task Force on Artificial Intelligence (YTAI). (2024, June). Report of the Yale Task Force on Artificial Intelligence: Reflections and Recommendations.
• Yale University. (2024, August 28). Yale Announces $150 Million to Support Leadership in AI. Yale News.
• Yale Poorvu Center for Teaching and Learning. (2024). AI Guidance for Students; Academic Integrity and AI Course Policies.
• Yale University. (2024). AI @ Yale: Guidance Hub & Knowledge and Tools Resources.

B. NUC, Government and Regulatory Documents

• National Universities Commission (NUC). (2024). Towards Responsible AI Integration in the Nigerian University System. NUC, Abuja.
• National Universities Commission (NUC). (2024). Roadmap for the Integration of Artificial Intelligence into the Nigerian University System. NUC, Abuja.
• UNESCO-ICHEI / IIOE / Ahmadu Bello University. (2024, July 18). Proceedings of the National Policy Dialogue on AI in Nigerian Higher Education. Abuja.
• Virtual Institute for Capacity Building in Higher Education, NUC, NOUN, TETFund, et al. (2025, February 17). Communiqué of the 2025 AI Awareness Day in Higher Education, Nigeria.
• Voice of Nigeria. (2026). Nigerian Government Urges Universities to Embrace AI, Accountability.
• Federal Ministry of Communications, Innovation and Digital Economy (FMCIDE). (2024). Nigeria National Artificial Intelligence Strategy 2024–2028.
• Federal Republic of Nigeria. (2023). Nigeria Data Protection Act 2023.
• National Information Technology Development Agency (NITDA). (2019). Nigeria Data Protection Regulation (NDPR).

C. Academic Journal Articles and Research Reports

• Cudjoe, D., et al. (2025). Investigating the existence of AI policy in higher education institutions in Sub-Saharan Africa. Proceedings of the Association for Information Science and Technology.
• Olakulehin, F.K., & Olojede, I. (2025). Influence of Generative Artificial Intelligence (GenAI) in Nigerian Higher Education. Àgìdìgbo: ABUAD Journal of the Humanities, 13(1), 228–252.
• Afolaranmi, A.O. (2025). Towards Formulation of Policies and Action Plans to Regulate the Use of Generative AI by Postgraduate Students in Educational Institutions in Nigeria. Zenodo.
• Ebhodaghe, I.U. (2025). Generative Artificial Intelligence and Academic Integrity in Nigerian Higher Education: Opportunities, Challenges, and Policy Implications. ResearchGate.
• Nnaji, I.L. et al. (2025). Enhancing Governance and Performance in Nigerian Higher Education through AI-Driven Digital Transformation. Social Sciences and Humanities Open.
• Eleje, L.I. et al. (2025). Artificial Intelligence Adoption in Higher Education in Nigeria. Discover Education. Springer.
• Ojedeji, S.O. (2026). Navigating Academic Integrity in the Age of On-Demand Artificial Intelligence. Frontiers in Education.
• Poudyal, B. (2025). Stress-Testing University AI Governance: A Prospective Method for Locating Policy Breakpoints. arXiv Preprint.
• Jiao, J. et al. (2024). The Global Landscape of Academic Guidelines for Generative AI and Large Language Models. arXiv Preprint.
• Di Monte, L., Appleby, M., & Hendrickson, D. (2024, December). AI at Yale: Updates on Institutional and Library GAI Initiatives. CNI.
• Yale Daily News. (2026). Yale Confronts AI with New Classes, Varied Policies and an Ethics Focus.
• Thesify.ai. (2026). Generative AI Policies at the World’s Top Universities: 2026 Update.
• Educause Review. (2024, November). Enabling Equitable Generative AI Access at Yale.
• Trinka.ai. (2024). Yale University AI Policy Repository.

D. International Policy Frameworks and Related Documents

• UNESCO. (2021). Recommendation on the Ethics of Artificial Intelligence. UNESCO, Paris.
• Blueprint.ng. (2025). Scholars Want Law on Comprehensive National AI Policy in Nigeria.
• Times Higher Education. (2026). THE Impact Rankings 2026.

---

This document is an original institutional policy review produced by Hallmark University for the purpose of AI policy development. It is a review and recommendation framework prepared ahead of the university’s forthcoming Senate-approved Artificial Intelligence Governance Policy, and does not itself constitute that final approved policy. Prepared October 2026, for NUC policy communication and for institutional and public use.