Policy on the Use of Artificial Intelligence (AI) in Student Assessment

This policy outlines the guidelines and principles for the responsible use of Artificial Intelligence (AI) applications, including generative AI, in assessment at Swansea University. The purpose of this policy is to promote ethical, transparent, and equitable use of AI, ensuring its potential benefits are harnessed while mitigating associated risks and challenges to maintaining standards.

This policy applies to all faculty staff, students, and professional service staff involved in assessment and/or the use of AI technologies within the University.

Evidence based advice and guidance will be shared among stakeholders. For students Artificial Intelligence Guidance - Swansea University, for staff Enhanced Artificial Intelligence (AI) guidance for staff - Swansea University.

The policy supports the University’s strategic vision and purpose Vision and ambition - Swansea University “enabling local solutions to the global challenges that affect us all”.

1. Swansea University will work in partnership with students, and support students and staff to become AI literate.
2. Swansea University trusts its students to use AI ethically and transparently.
3. Staff will be equipped to support students to use generative AI tools effectively and appropriately in their learning experience.
4. Swansea University will adapt teaching and assessment to incorporate the ethical use of generative AI and support equal access.
5. Swansea University will ensure academic rigour, integrity and standards are upheld.
6. Swansea University will work collaboratively to share best practice as the technology and its application in education evolves.

(adopted from the Russell Group Principles on the use of generative AI tools in education)

i. All users of AI applications within the University must adhere to ethical principles, respecting human rights, privacy, and data protection.

ii. The use of AI should align with the University's mission, values, and academic goals, supporting teaching, learning, research, and administrative functions.

iii. Approaches involving the use of AI systems should not discriminate against individuals or groups based on characteristics protected by law, such as race, gender, ethnicity, religion, disability, or socioeconomic status.

iv. Staff and students will not be permitted to present work completed by AI applications as their own without acknowledgement.

i. All AI applications used in the University should be transparent and accessible to all. Users should have a clear understanding of how AI systems make decisions and the underlying algorithms used, including the risks, limitations and biases within said algorithms.

ii. Inaccuracy of information: generative AI tools gather information from a wide variety of sources, some of which are incorrect, misinformed, or incomplete. The inputs and commands by users may be unclear or inaccurate. This therefore means that results and outputs produced by generative AI tools may be inaccurate, irrelevant is misleading. Users must always ensure the accuracy of results produced by generative AI tools.

iii. There will be active monitoring and mitigation of bias in AI systems and their usage within the University to ensure fairness and prevent discrimination.

iv. The University will provide necessary accommodations and support to ensure equitable access to AI-based resources and services.

i. Use of any AI applications must comply with relevant General Data Protection Regulations (GDPR).

ii. Staff and students should not enter any identifiable information into any AI application.

iii. Any data collected and/or analysed by AI systems should be obtained with appropriate consent, ensuring privacy and confidentiality.

iv. Staff or students are not permitted to upload student assessment, or any identifiable data without explicit written consent, to AI applications for any reason (including attempts to check for AI generated content).

i. The University will enable the use of AI applications in assessment where this is appropriate and relevant to enhance student learning.

ii. Assessment will be designed to ensure that integrity and standards are maintained where students have access to generative AI, without sacrificing the importance of authenticity of assessment and pedagogic practice.

iii. The Assessment, Marking and Feedback Policy, Proofreading Policy and Academic Misconduct procedure will make reference to appropriate usage of AI in assessment, explaining the risks and consequences of inappropriate usage for students.

i. Intellectual property rights related to work generated wholly or in part by artificial intelligence will be clarified and defined within the University’s IP Policies.

Training, support and professional development opportunities for staff and students involved in the development, deployment, or use of AI technologies will be provided by the University.

Training should focus on promoting effective usage, awareness of ethical considerations, data privacy, algorithmic bias, and other relevant topics.

Non-compliance with this policy will be reported to the Regulations, Quality and Standards Committee.

In the immediate term, given the pace of change in this area, it is likely that this policy will be updated as frequently as necessary to maintain standards.

Formally, this Policy will be reviewed at least annually.