Artificial intelligence records
Artificial intelligence and recordkeeping
Artificial intelligence (AI) is becoming an increasingly common tool to support productivity and discovery. Use of AI is guided at the University by a set of AI principles, a key principle being the need for appropriate human oversight when using AI. As with any other business activities, our use of AI needs to be transparent and accountable.
When using AI tools (eg, SparkAI, Copilot) to support your work activities, it’s important to understand your obligations around managing AI generated content as records.
When does AI content become a record?
When AI is used to generate content and inform decision-making that supports and evidences University activities, these are University records. All staff have an obligation to create and keep records of their work activities, including content generated through AI, under the Records Management Policy (MPF1106).
AI records management considerations
If you’re using AI to support the work you do, here are key records management requirements to consider:
| Generated content |
|
| Decision making |
|
| Process automation |
|
| Storing outputs |
|
| Disposing of AI data |
|
Resources
If you want to learn more about AI use, additional informational resources are included below:
- University of Melbourne AI principles
- Using Generative Artificial Intelligence Safely and Effectively: Guardrails for Professional Staff at The University of Melbourne
- Statement on responsible use of digital assistance tools in research
- Assessment, AI and Academic Integrity
- Managing privacy risks when using Artificial Intelligence tools KBA
Further information
If you require further information or advice on records disposal, please submit a ServiceNow request.