In the global education sector, we have spent much of the last two years looking at artificial intelligence through a defensive lens. The conversation has been dominated by concerns over academic integrity and the perceived threat to traditional assessment. However, as we look at the challenges facing UK higher education in 2026 – chief among them student retention and the widening attainment gap – it is time to shift our perspective.
What if AI isn’t the problem, but a vital part of the solution?
New independent research by Dr Rebecca Mace SFHEA, titled Ethical AI in Higher Education: Boosting Learning, Retention and Progression, provides a data-backed argument for this shift. By analyzing over 8,000 data points from diverse UK institutions, the report reveals that when AI is used as a formative “learning partner,” it creates a “stabilising scaffold” that keeps students in school and helps them thrive.
The “equalising effect” on attainment
For international and domestic students alike, the leap to university-level academic writing can be daunting. Dr Mace’s research found that formative AI feedback has a powerful “equalising effect”. While writing scores improved across the board, the most rapid gains were seen among lower-performing students.
The research tracked measurable improvements in core academic areas for students using Studiosity AI for learning:
- Text analysis: +10.98 points
- Scientific reports: +7.18 points
- Essays: +6.72 points
This isn’t about AI writing for the student; it’s about the student using feedback to master “academic code-switching” – the ability to translate their ideas into the formal language of their discipline.
A roadmap for retention
Retention is the “holy grail” for university leaders today. The Mace report identifies a direct positive correlation between the use of Studiosity formative AI for learning and student persistence.
The data suggests that learning is an iterative process. Students who engaged with the tool showed consistent progress over time, with six submissions appearing to be the “sweet spot” where academic standards become internalised. For a student who might be struggling in silence at 2:00am, having an ethical, 24/7 feedback loop provides the confidence to keep going rather than dropping out.
From guilt to growth
Perhaps the most revealing part of the study is the psychological impact on students. Many reported feeling a sense of “guilt” when using AI, even for legitimate study support, due to a lack of clear institutional guidance. This “low-trust culture” is counterproductive.
As university leaders, you have an opportunity to validate ethical AI use. By providing students with approved, pedagogy-first tools, we move them away from the “gray areas” of the internet and back into a structured, supported learning environment.
Take the next step
The evidence is clear: ethical AI is no longer a luxury or a risk to be managed – it is a strategic necessity for any institution serious about student success and social mobility. I invite my colleagues across the sector to dive into the data and see how these findings can be applied to your own student success strategies.
About the author: Isabelle Bambury is the managing director UK and Europe at Studiosity. Isabelle has over 20 years’ experience in the education sector, before Studiosity as regional director for Study Group where she led both the UK/Europe and Russia/Central Asia teams. Prior to this, Isabelle held key roles at Cambridge Education Group and Kaplan International, moving into the private sector in 2005 after beginning her career as a secondary school teacher.
The full report is available for download at www.studiosity.com/download-ethical-ai-studiosity
About Studiosity: Support and Validate. Studiosity is writing feedback and assessment security that helps educators and leadership support students and validate learning outcomes, and unlike police & punish detection technology, Studiosity helps protect degree value, pedagogically and ethically.
