In March, HEPI published its third annual student survey on generative AI. Ninety-five per cent of undergraduates now use AI in some form, and 94% use it on assessed work. Only 36% feel their institution encourages them to, and only 38% say their institution provides them with the tools. (HEPI, Student Generative AI Survey 2026)
That is usually read as a policy lag. I think it is something else. It is an equity decision that institutions are making by not making it.
When a university does not provide the tool, provision does not stop. It moves to the student. Access then depends on who can pay twenty pounds a month, who has a friend in a computing course, and who has the confidence to ask a tutor whether they are allowed. The same survey found that students at Russell Group institutions are the most likely to say their institution encourages AI use, up from 26% to 39% in a year. The sector’s AI divide is forming along the lines of advantage that already exist, and it is forming quickly.
Staff and organisational developers are holding this, largely without a budget for it. You are being asked to run AI literacy sessions, interpret guidance you did not write, and reassure colleagues who suspect the technology is coming for their job. It is a lot to carry, and most of the sector’s AI conversation is not helping you, because it is a conversation about tools.
I have spent the past year interviewing people who use AI deliberately to close equity gaps rather than to trim admin. Dr Pauldy Otermans, an academic who also runs Otermans Institute, builds AI teaching avatars to reach learners that formal education has never served well, including displaced people and adults with low literacy. Anna Jones at RefuAid works with refugees whose qualifications are not recognised here, so that a Syrian doctor is not driving a delivery van and actually utilising their skills. Neither of them is doing something a university could not do. They are simply doing it on purpose.
That is the finding I did not expect. The technology is not the constraint. The pioneers have the intention, institutions have the power to scale, and they are rarely the same people.
Across twelve of those conversations, these seven commitments form the EQUITAS framework, which Leah Sunshine and I explore in more detail in our forthcoming book, AI for Equity: Creating a More Equitable Society for All.
Examine whose world the data describes. Every tool you procure was trained on someone’s idea of a normal student. Ask your supplier whose.
Act with, not for. If your AI guidance was written for widening participation students rather than with them, you have designed for a student who does not exist.
Start before you are ready. Waiting for institutional policy is itself a decision, and it is the one with the clearest losers.
None of this needs a new committee. It needs the people who design development in universities to treat AI as an equity question rather than an IT one, and to say so out loud while the norms are still soft.
So here is the question I would put to every staff developer reading this. Who in your institution is currently paying for their own AI access out of their own pocket, and what does the answer tell you about who your provision is really for? You will already know the answer for your staff. The harder version is your students.
Ninety-five per cent of them are already using it. Deciding which of them it works for is still within your gift, but not for much longer.
Testimonials
Confirmed:
“AI for Equity is a collective call to action to resist the current trajectory of AI and join those making technology which serves everyone equitably. Jenny and Leah-Sunshine Garrett’s conversations with experts reveal that a future with AI does not have to be feared; instead, it is an opportunity to create a tool which reflects humanity not as it is, but as it could be.”
Kyra Wilson, University of Washington
“Most books about AI ask you to think differently. This one asks you to act. The most radical thing it does is also the simplest: it listens to practitioners, communities, and people whose knowledge has been sitting outside the room where AI decisions get made. Jenny and Leah-Sunshine Garrett have written the book the field has been waiting for without knowing it.”
Karen Blake MBE Hon. FBAASc, former Co-CEO, Tech Talent Charter; Honorary Fellow of the British Science Association
Jenny Garrett OBE is Patron of the Staff Development Forum and CEO of Jenny Garrett Global. AI for Equity: Creating a More Equitable Society for All*, co-authored with Leah-Sunshine Garrett, is published by Emerald on 27 October 2026 (GBP19.99, ISBN 978-1-83742-857-1) and is available to pre-order at jennygarrett.global/ai-for-equity.