Ironic Integrity

How Turnitin plans to use our data to train AI

That a single announcement can impress upon an entire mass of students is a bitterly remarkable feat. It would not be out of place to liken the actions of Turnitin to those of Marcus Brutus or Fredo Corleone, with their proposal to hand students’ work over to the training of artificial intelligence systems. The platform, on whose function 98% of universities in the United Kingdom rely on to detect plagiarism and AI-generated work, intends to use the intellectual property of students to promote the very technologies enabling such indolence.

On 14 July 2025, Turnitin proposed changes to its end-user license agreement (EULA), which allowed the use of student submissions to improve and develop AI tools. This prompted an immediate uproar across higher education institutions, stirring up obvious concerns regarding data privacy and contractual rights, according to Times Higher Education. The largely united response proved enough to halt the changes, if only temporarily. As it stands, Turnitin has agreed to postpone the issue of a new EULA until September 2027.

In an Instagram post, the University of St Andrews Students’ Association’s position was clear: “Tertiary education depends on original thought, intellectual ownership, and critical inquiry” and Turnitin’s proposal “risks undermining these principles.” Calling for governmental  intervention, they argued for the students' control over how their intellectual property is used, asserting the responsibility is incumbent upon Turnitin  “to uphold the interests of the learners they serve.” Their official response to Turnitin, supported by 86 fellow student office representatives from across the nation, has three main points:

1. Turnitin must permanently withdraw its proposed changes and commit to protecting student work from use in the training or development of AI technologies.

2. Turnitin must engage directly with student leaders before pursuing any further changes relating to AI, student data, intellectual property, or academic integrity.

3. The UK and all devolved governments will prevent any tertiary institution from supporting academic platforms that harvest students’ work to enhance AI systems.

The petition has swiftly garnered attention, having already risen from a reported 67 signatories in under a week; a testament to the ethical convictions and united concerns of our representatives. However, the new EULA may still come into effect, and if so, what options remain for the opposing universities? Turnitin has risen to dominate the educational system across the world, containing a repository of over 1.6 billion submitted papers, as well as millions of scholarly resource publications and journals. These sources provide an extensive database for comparison, acting as a complement to AI-generated prose estimates. Turnitin is also capable of granting greater scope for understanding the authenticity of work by returning estimates in percentage bands, unlike most other tools which use only binary verification. Considering such points, it will clearly take a great deal for a new platform to prove capable of functioning as an equally effective replacement.

But as students — the ones at greatest threat of redundancy in this new artificial age — are we predisposed to reject contributions to the advancement of AI technology? Do the unchanged intentions of Turnitin, despite inciting such negative responses, indicate any benefits to their proposal? In any case, it is difficult to reconcile the idea of a tool defending intellectual integrity — above all, when that very tool is produced with the intent to replace thought itself.  

Image from Wikimedia Commons