Faculty/module/cohort:
Marketing modules across undergraduate (Levels 4–6) and postgraduate programmes, including a final-year live client project and a Professional Practice module.
AI Tools Used:
ChatGPT and comparable generative AI chatbots; AI image-generation tools for campaign visualisation; AI voice-interaction tools for interview preparation; a school-specific AI dissertation-support agent.
AI Modes Used:
- Text → Text (research support, drafting, feedback)
- Text → Image (campaign visualisation and marketing assets)
- Voice Interaction (AI-generated interview simulation for employability preparation)
What was the challenge?
The first concerned marketing assessments that required students to develop visual campaign concepts. While many students generated strong creative ideas, some lacked the design skills needed to communicate those ideas effectively through professional-quality marketing assets.
The second related to a research-intensive live client project delivered within a thirteen-week semester. The original assessment structure placed considerable demands on students, with approximately half of one previous cohort requesting assessment extensions. This prompted a review of the assessment design and consideration of how AI could be integrated to better support the learning process.
What did they do?
A range of AI-supported practices that have been incorporated into marketing teaching over several years. These include AI image-generation tools to enable students to visualise marketing campaigns and create campaign assets, allowing design software proficiency to become less of a limiting factor when communicating marketing ideas.
Within the live client module, the research assessment was redesigned as a Category 3 (AI-integrated) assessment. Students were expected to use AI to support the analysis of relationships and strategic tensions between three real client organisations while demonstrating critical evaluation of the outputs.
To promote transparency, students completing Category 2 and Category 3 assessments were required to include a specific declaration identifying both the AI tool used and the purpose for which it had been used, rather than submitting a general statement that AI had assisted the work.
To give another example, AI voice-interaction tools were introduced within a Professional Practice module. Students provided contextual information about forthcoming industry networking events, after which AI generated potential interview questions that students answered through spoken interaction before attending the real event.
Beyond assessment, an AI dissertation-support agent was trained using module documentation and teaching materials to provide students with consistent guidance alongside supervision.
AI was also used to help draft supportive written feedback for a student with formally recognised additional learning needs and to provide individual guidance to a mature international student who had limited prior experience with AI before undertaking an AI-integrated assessment.
Multimodal Use
This approach to practice incorporates multiple AI modalities to support different aspects of teaching and learning.
Text-based AI supports research, drafting and feedback processes. Image-generation tools enable students to visualise campaign concepts and produce marketing assets. Voice-interaction tools provide opportunities for students to rehearse professional conversations through simulated interview scenarios.
Rather than serving a single purpose, each modality was described as addressing a different aspect of learning or assessment, including idea development, communication, employability preparation and academic support.
What did this look like in practice?
Students used AI image-generation tools to produce campaign visuals and marketing assets for assessed work, allowing visual prototypes to accompany written campaign proposals.
In a disability-focused marketing project AI-generated images produced inaccurate representations of disabled people. Rather than removing the example, it has subsequently been used within teaching to encourage discussion about representational bias in generative AI.
Similarly, a publicly available example in which an AI-generated family photograph inserted a male figure into an image of a lesbian couple with their child was used to encourage students to critically evaluate AI outputs rather than accepting them uncritically.
Other examples include the use of mandatory AI acknowledgment statements within assessments and AI-supported interview simulations undertaken before industry networking events.
What was the impact?
There were several changes following the introduction of AI-integrated assessment.
The revised Category 3 assessment achieved a 100% submission rate without extension requests, compared with a previous cohort in which approximately half of the students had requested extensions under the earlier assessment structure. The submitted work demonstrates greater depth of analysis and subsequent classroom discussions suggested students had engaged more thoroughly with the underlying material. However, these observations reflect professional judgement rather than a formal evaluation study.
Differences were observed in how students appeared to engage with AI. In his experience, higher-performing students were more likely to use AI as a starting point for further critical thinking, whereas students who relied more heavily on AI outputs without further evaluation tended to produce weaker work. This suggests that the educational value of AI depends substantially on how students engage with it.
Some students appeared more willing to discuss their AI use openly than in previous years. This could be attributed, in part, to explicit messaging within the module that transparent and responsible AI use was expected where appropriate.
Ethical or practical considerations
Several practical and ethical issues arise from AI use in higher education:
Bias and representation
AI-generated images can reinforce stereotypical or inaccurate representations, including distorted depictions of disabled people and assumptions about family structures. Examples can be used as teaching resources to encourage critical evaluation of AI-generated content.
Academic integrity
AI has introduced new challenges for academic integrity. There can be a disconnect between a student’s classroom engagement and the sophistication of submitted work, although these situations remain matters for academic judgement rather than definitive indicators of AI misuse.
Digital inclusion
Students have differing levels of prior AI experience, for example additional guidance was needed for a mature international student unfamiliar with AI before undertaking an assessment in which AI use formed an expected component.
Institutional practice
AI may have potential to support aspects of assessment, such as acting as a supplementary reviewer alongside academic judgement.
Reflections and Advice
What worked well?
Two practices were particularly effective: requiring students to provide specific AI acknowledgements that identify both the tool used and its purpose and using authentic examples of AI bias as opportunities for classroom discussion.
What would you refine?
Further work is needed to help educators identify situations where students may be relying on AI uncritically while avoiding approaches that unnecessarily discourage legitimate AI use.
Advice for colleagues
AI can be presented as an extension of digital literacy rather than as a separate or exceptional technology. Tutors can discuss openly how AI is being adopted within professional practice so that expectations surrounding responsible use remain transparent.
Further reflections
Institutional attitudes towards AI continue to evolve. Ongoing dialogue should focus less on whether AI should be used and more on how it can be integrated in ways that support learning while maintaining academic standards.
Quick Start Guide for Colleagues
Estimated time required:
Low to Moderate. Much of the work involves redesigning existing assessment structures rather than creating entirely new activities.
Digital skill level required:
- Students: Low to Moderate
- Staff: Moderate, particularly where AI use is incorporated into assessment design and guidance.
Common pitfalls to avoid:
- Requiring generic AI disclosure statements rather than asking students to explain how AI contributed to their work.
- Assuming all students possess comparable levels of AI literacy or prior experience.
- Treating AI outputs as authoritative without encouraging critical evaluation.
Best suited for:
Marketing and business education, employability-focused modules, live client projects, and other assessments combining written, visual and professional communication outputs.
