The challenge of AI in assessment

Assessments allow us to measure and evaluate student knowledge and competencies in a subject discipline, often by assessing a product or artefact. These products and artefacts are commonly created in written form, whether that is an essay, exam, report, poster presentation or reflection. However, Generative AI poses a significant threat to the integrity of these assessments. It can hinder our ability to accurately and confidently determine whether students have met their intended learning outcomes, and are prepared to become safe and competent professionals.

An image reading "53% of students have used generative AI to help them with assessments". Statistic taken from HEPI, "Provide or punish? Students’ views on generative AI in higher education", February 2024

Awards at Newcastle University are made, and classified, based on evidence that students have met or exceeded the learning outcomes for their programme of study. However, as stated by the QAA:

"the rapid rise and ubiquity of Generative Artificial Intelligence software means that some or all the assessments that currently contribute to the evidence base may no longer be confidently ascribed to an individual student."

This situation will only become more challenging as the AI platforms and tools evolve and become embedded in our everyday lives.

How big is the problem really?

As evidenced in a recent HEPI poll of 1,250 UK undergraduate students (2024), more than half of the students surveyed (53%) used generative AI to help them with assessments. More than one-in-eight (13%) used AI to generate text for their assessments, but they typically edited the content before submitting. One in 20 (5%) admitted they simply copied-and-pasted AI-generated text without editing it. Although high in their own right, we believe these numbers to be extremely conservative!

Student understanding of how to use AI - and more so how to daisy-chain tools together to produce more effective outputs that are impossible to detect - is also growing. A Newcastle University AI consultant has been building a list of TikTok videos of students advocating AI tools to avoid detection. Disclaimer - prepared to be shocked!

The need for change

In line with Education for Life 2030+ and 5 Principles for the Use of AI, we need to review our assessments to ensure they meet intended learning outcomes and maintain academic integrity. We should also aim to reduce the emphasis on high-stakes exams, favouring more diverse, authentic, and lower-weighted assessments – assessments that offer students better learning opportunities, encourage them to improve through feedback, and which enhance their overall perception of the assessments' value.

On the next page we will begin to consider the approaches we can take to reviewing your current assessment approaches.