A Practical Framework for Ethical AI Use in Higher Education
Generative AI has changed how students find information, analyse ideas, develop arguments and produce academic work. The Five-Stage Framework provides a structured way to use AI without surrendering intellectual ownership, academic judgement or accountability.
The principle is simple: AI may assist the process, but the student remains responsible for the academic work.
Why a Framework Is Needed
Traditional approaches to academic integrity were largely developed around questions of plagiarism, authorship, unauthorised collaboration and the improper use of sources. Generative AI introduces a different challenge because it can contribute at many points in the academic process—not simply to the final written submission.
Students can now use AI to explore a subject, analyse information, develop connections between ideas, generate material and assist with the presentation of academic work. A simple distinction between “AI used” and “AI not used” therefore tells us very little about the student’s actual intellectual contribution.
The important question becomes: How can students benefit from AI while remaining intellectually responsible for the work they submit?
The Five-Stage Framework focuses not simply on whether AI was used, but on how it was used, where human judgement was exercised, and whether the student can demonstrate intellectual ownership of the final work.
The Five Stages
The framework follows the progression of academic work from initial exploration to transparent documentation.
01
EXPLORE
Use AI to broaden enquiry, generate possibilities and identify avenues for investigation—without treating AI output as established fact or a substitute for independent research.
Human responsibility:
Question, verify and decide what deserves further investigation.
02
ANALYSE
Evaluate evidence, arguments and AI-generated material critically. Test assumptions, identify weaknesses and compare AI outputs with credible academic sources.
Human responsibility:
Judge quality, relevance, accuracy and credibility.
03
SYNTHESIZE
Bring together evidence, disciplinary knowledge, competing perspectives and your own developing understanding into a coherent intellectual position.
Human responsibility:
Make the connections and determine what the evidence means.
04
CREATE
Develop the argument, solution, interpretation or academic output that represents your own intellectual contribution. AI may assist, but should not replace the thinking the assessment is intended to demonstrate.
Human responsibility:
Own the argument, decisions and final academic product.
05
DOCUMENT
Record and disclose material AI involvement clearly enough for the academic process to remain transparent and accountable.
Human responsibility:
Explain how AI contributed and take responsibility for the submitted work.
How the Five Stages Work
Each stage identifies both legitimate opportunities for AI assistance and the point at which human academic responsibility must remain decisive.
01
EXPLORE
Opening the enquiry
What happens at this stage?
Students may use AI to clarify terminology, generate initial questions, identify possible approaches, obtain preliminary orientation and consider perspectives that might warrant further investigation.
The academic-integrity risk
AI output can sound authoritative even when it is incomplete, inaccurate or unsupported. Exploration becomes problematic when preliminary AI output is treated as evidence or replaces independent research.
The student’s responsibility
The student decides what is worth pursuing, verifies claims through credible sources and determines the direction of the enquiry.
02
ANALYSE
Testing claims and evidence
What happens at this stage?
Students interrogate evidence, compare sources, test assumptions and evaluate AI-generated suggestions rather than accepting them at face value.
The academic-integrity risk
AI can produce persuasive analysis without reliable evidence, conceal uncertainty or reproduce weaknesses in the information on which it was trained.
The student’s responsibility
The student must judge credibility, relevance, accuracy and significance.
03
SYNTHESIZE
Connecting evidence and ideas
What happens at this stage?
Students bring together evidence from different sources, perspectives and analyses to develop an integrated understanding of the problem.
The academic-integrity risk
Combining AI-generated summaries or passages is not the same as intellectual synthesis. Genuine synthesis requires judgement about relationships, significance, disagreement and meaning.
Ethical AI use is considered throughout the academic process—not merely by examining the final submission.
The student’s responsibility
The student establishes the connections, reconciles competing evidence and determines the intellectual position that emerges.
04
CREATE
Making the intellectual contribution
What happens at this stage?
The student transforms exploration, analysis and synthesis into an argument, interpretation, recommendation, design, solution, report or other assessed output.
The academic-integrity risk
Excessive AI generation at this stage can obscure whose intellectual work the final submission actually represents.
The student’s responsibility
The student must own the argument, decisions and academic contribution represented by the submission.
05
DOCUMENT
Making AI involvement transparent
What happens at this stage?
The student records material AI involvement and discloses it in accordance with the requirements of the assessment and institution.
The academic-integrity risk
Undocumented AI involvement makes it difficult to distinguish legitimate assistance from inappropriate substitution of student work.
The student’s responsibility
The student should be able to explain where AI contributed, how its output was evaluated and which academic decisions remained their own.
Documentation turns responsible AI use into a process that can be explained and evidenced.
What Makes the Five-Stage Framework Different?
The framework is built around four principles that keep human judgement, intellectual ownership and accountability at the centre of academic work.
01
Intellectual Ownership
Students remain responsible for the reasoning, judgement and academic contribution represented by their work.
02
Process, Not Just Product
Ethical AI use is considered throughout the academic process—not merely by examining the final submission.
03
Transparency
Material AI involvement should be recorded and disclosed appropriately.
04
Human Accountability
AI may generate content, suggestions or analysis, but responsibility for academic decisions remains human.
AI can assist academic work. It cannot assume academic responsibility.
From Framework to Practice
The Five-Stage Framework can be understood, practised and implemented through three complementary routes.
UNDERSTAND
The Book
Explore the rationale, principles, academic-integrity implications and complete Five-Stage Framework.
APPLY
The Workbook
Put the framework into practice through guided activities, cases, reflection prompts and documentation tools.
IMPLEMENT
Consultancy
Bring the framework into institutional policy, teaching, assessment and staff development.
A Shared Framework for Students, Staff and Institutions
FOR STUDENTS
Use AI Without Losing Ownership of Your Work
The framework helps students ask five important questions throughout their academic work:
Where can AI assist me?
What thinking must remain mine?
What must I verify?
What should I disclose?
Can I explain and defend my final work?
FOR STAFF & INSTITUTIONS
Create a Consistent Approach to Responsible AI Use
The framework provides a common basis for considering:
Policy
Teaching
Assessment
AI literacy
Documentation
Academic-integrity decisions
Ready to Put the Framework into Practice?
Whether you are a student developing responsible AI practice, an educator integrating AI into teaching and assessment, or an institution developing a coherent approach to AI and academic integrity, the Five-Stage Framework provides a practical starting point.
For institutional implementation → Explore Consultancy