library(tidyverse)
library(plotly)
library(readxl)
library(knitr)Generative AI has become nearly universal among university students. While students increasingly rely on AI tools for learning, research and productivity, universities continue adapting policies, assessments and guidance. This story explores how students use AI, what forms of AI assistance they consider acceptable, and whether institutions are keeping pace with this rapid technological shift.
Students are using AI tools across a wide range of academic activities, with text-generation applications emerging as the dominant use case.
More than half of respondents reported using text-generation tools such as ChatGPT, making it the most widely used form of AI among students. Other common uses include textbook summarisation, writing enhancement and translation support. In contrast, fewer students reported using AI for coding, data analysis or companionship purposes.
These results suggest that students primarily view AI as a tool for improving productivity and supporting learning tasks rather than replacing human interaction.
Student attitudes towards AI vary considerably depending on how the technology is used.
Across all subject areas, students were most accepting of AI being used to explain concepts, summarise articles and generate research ideas. Acceptance declined substantially when AI was used to generate assessment content directly, particularly when students submitted AI-generated work with little or no editing.
This pattern suggests that students generally view AI as an acceptable learning aid but remain cautious about uses that could undermine academic integrity.
Most students believe their institutions are providing guidance on the use of AI in academic work.
Positive responses, including both Agree and Strongly Agree, account for the majority of responses across all subject areas. Health students reported the highest levels of institutional guidance, while Arts and Humanities students were slightly less likely to report receiving clear guidance. Negative responses remained relatively uncommon across all disciplines.
These findings indicate that universities have begun responding to the rapid adoption of AI by introducing policies and guidance for students.
Students increasingly perceive that AI is influencing how universities design and assess coursework.
Agreement that assessment practices have changed because of AI was highest among STEM and Social Science students, where positive responses substantially outweighed negative responses. Although perceptions varied slightly across disciplines, most subject areas reported more agreement than disagreement. Only a small minority of students strongly disagreed that assessment had changed.
This suggests that the growth of generative AI is already affecting assessment design and encouraging institutions to adapt existing approaches.
Acceptance of AI-generated work varies noticeably between academic disciplines.
Arts and Humanities students were more likely to disagree with the inclusion of AI-generated content in assessed work, while Health, STEM and Social Science students displayed higher levels of acceptance. Despite these differences, neutral responses remained common across all disciplines, suggesting ongoing uncertainty about the role of AI in academic assessment.
The results highlight that attitudes towards AI are influenced by disciplinary norms and expectations, meaning a single institutional policy may not reflect the views of all students.
Generative AI has rapidly become part of everyday student life, with many students using AI tools to support learning, research and productivity. While students generally accept AI as a learning aid, they remain cautious about uses that directly affect assessment integrity. Universities appear to be responding through policy development and assessment redesign, although attitudes towards AI continue to vary across disciplines. As AI technologies continue to evolve, universities will need to balance innovation, assessment integrity and student expectations when developing future policies and teaching practices.
Data used in this story was sourced from the Higher Education Policy Institute (HEPI) Student Generative Artificial Intelligence Survey 2026. The survey collected responses from 1,054 full-time undergraduate students studying in the United Kingdom.
Relevant survey tables were extracted from the published dataset and transformed into analysis-ready data frames. Visualisations were created in R using ggplot2 and plotly to support interactive exploration of trends in AI use, institutional policy, assessment changes and disciplinary differences in attitudes towards AI.
ChatGPT was used during the development of this assignment to assist with brainstorming visualisation ideas, debugging R code, understanding survey table structures and improving written explanations. All analytical decisions, chart construction, interpretation of results and final written content were reviewed, verified and edited by the author.
Higher Education Policy Institute (HEPI). (2026). Student Generative Artificial Intelligence Survey 2026.
Stephenson, R., & Armstrong, C. (2026). Student Generative Artificial Intelligence Survey 2026. Higher Education Policy Institute (HEPI).
https://www.hepi.ac.uk/reports/student-generative-ai-survey-2026/