More students getting caught using Ai while plagiarism declines
While the incidences of plagiarism seem to be on the decline, more students are getting caught using Ai dishonestly for their academic writing. Here we take a look at the latest developments.
Sharp rise in AI‑related cheating
In the 2023–24 academic year, UK universities reported almost 7,000 confirmed cases of cheating using AI tools like ChatGPT—equivalent to 5.1 cases per 1,000 students, a significant jump from 1.6 per 1,000 the previous year. Early data for the current academic year suggests this could rise further to 7.5 per 1,000, though experts stress that these officially recorded cases are likely just the tip of the iceberg (theguardian.com).
Traditional plagiarism on the decline
While AI misuse surges, conventional plagiarism is decreasing—from 19 cases per 1,000 students in 2019–20 to 15.2 in 2023–24, with projections indicating it may fall to around 8.5 per 1,000 (theguardian.com).
Data from universities
Under Freedom of Information requests, The Guardian reached out to 155 universities; 131 responded. Over 27% of these institutions did not yet categorise AI‑based misconduct separately in 2023–24, demonstrating how ill‑prepared many are to handle this shift (theguardian.com).
Detection challenges & the “tip of the iceberg”
Detecting AI authorship poses unique difficulties compared to traditional plagiarism. There’s no direct text to compare, and faculty often hesitate to falsely accuse students. The University of Reading, for example, found 94% of AI‑generated submissions evaded detection, reinforcing fears that most cases go unnoticed (theguardian.com).
Student behaviours & tools
AI aids like ChatGPT can be used legitimately—for brainstorming, summarisation, even supporting students with dyslexia. Yet many students exploit these tools covertly, running AI outputs through paraphrasing apps or “humanising” systems to bypass detectors. A student quoted noted, “ChatGPT … to help brainstorm … Anything that I would take … I would then rework completely in my own ways” (theguardian.com).
Academic perspective & assessment redesign
Experts warn that simply tightening sanctions or reverting to in‑person assessments is insufficient. Educators must redesign evaluation to focus on non‑replicable skills such as oral presentations, critical thinking, and teamwork. Involving students in designing assessments can foster buy‑in and deter misuse (theguardian.com).
Government response
The UK government has allocated over £187 million toward national skills programmes and AI guidance, aiming to integrate AI responsibly into education—mitigating misuse while supporting learning inclusively (theguardian.com).
Voices from campuses
A music lecturer shared that about half of his students likely use AI—knowingly or unknowingly—with some unaware that this counts as cheating (theguardian.com). Another Higher Education Policy Institute study found 92% of UK students used generative AI in their studies, sparking calls for universities to “stress-test” assessments (theguardian.com).
???? Key takeaways
- AI‑cheating is skyrocketing, with confirmed cases increasing more than threefold in a year.
- Traditional plagiarism is decreasing, but is being replaced by AI‑enabled misconduct.
- Detection methods are inadequate—many AI‐generated assignments go unnoticed.
- Assessment overhaul is needed: focus on interpersonal, reflective, and real‑time tasks.
- Government investment indicates a broader national push to embed AI into education carefully.
- Student and staff experiences vary—some use AI supportively, others exploit it sneakily.
In essence, the UK’s higher education system is at a pivotal crossroads: the rise of AI demands not just defensive tactics, but a reimagining of how we assess learning. While detection tools are evolving, the more lasting solution may lie in redesigning assessments to prioritise authentic human skills—with institutional and governmental support shaping the path forward.
