Red X iconGreen tick iconYellow tick icon
Skip to main content
Category Information Technology
Type Policy
Approved by Vice-Chancellor, 9 March 2026
Date Policy Takes Effect 20 May 2026
Last approved revision 
Sponsor Vice-Chancellor
Responsible officer Chief Digital Officer

Purpose

This Policy establishes clear expectations and requirements for the deployment and use of Generative AI Systems across all Ōtākou Whakaihu Waka | University of Otago activities. Specifically, this policy aims to:

  • Enable the use of Generative AI systems to enhance research, innovation, operational excellence and service delivery in alignment with the University's strategic objectives
  • Protect the rights, privacy, and well-being of students, staff, and stakeholders
  • Ensure transparency and accountability in AI practices to maintain public trust and confidence
  • Provide a clear framework for identifying, assessing, and managing risks associated with AI Systems
  • Support staff capability development and responsible AI literacy
  • Ensure compliance with legal obligations, university policies, and ethical standards.

Organisational scope

This policy applies to all University of Otago staff members, contractors and consultants when acting on behalf of the University.

Definitions

Advanced Technology Reference Group
A University committee advising the University Senior Leadership Team (SLT) about Advanced Technology, including AI and is responsible for monitoring future and current advanced technologies that impact the University's vision and mission.
Artificial Intelligence (AI) Systems
Any technology or tool that uses data to make inferences and generate outputs such as predictions, recommendations, decisions, or content (text, images, video, code) with a degree of autonomy. This includes machine learning models, generative AI tools, predictive analytics Systems, neural networks, natural language processing designed to enhance efficiency, decision making and automation, and chatbots that generate their own responses.
Generative AI
A type of AI that creates new content, including text, images, videos, and code, by learning from large amounts of data and generating responses to user prompts. Examples include ChatGPT, Claude, Microsoft Copilot, and similar tools.
Data Classification
The categorisation of information based on sensitivity level (Public, Internal, Confidential, Restricted) as defined in the Information Security Classifications Guidelines.
Deploy
To spread a system or service across the University and make it available to users at scale. This does not include AI systems developed and operated locally for teaching or research that are not intended for University-wide use.
Māori Data Sovereignty
Refers to the inherent rights and interests that Māori have in relation to the collection, ownership, and application of Māori data (further guidance is available from Te Mana Raraunga).
Māori Data
Refers to digital or digitisable information or knowledge that is about or from Māori people, language, culture resources or environments (further guidance is available from Te Mana Raraunga).

Content

  1. Context for use of AI Systems

    1. The University is committed to the responsible, ethical and effective use of Artificial Intelligence (AI) Systems. The adoption and use of AI Systems will support our Vision 2040, Pae Tata: Strategic Plan to 2030, uphold Te Tiriti o Waitangi, protect Māori data sovereignty and Pacific data sovereignty, and maintain the highest standards of integrity, transparency and accountability, in accordance with the AI Governance Policy.
    2. This Policy applies only to the use of AI Systems with generative AI or AI learning capabilities.  If in doubt, whether a technology is covered by this Policy should be determined by consulting the Chief Digital Officer, who may in turn consult the Advanced Technology Reference Group (ATRG).
    3. This Policy is to be read in conjunction with the Use of Generative-Artificial Intelligences and Autonomous Content Generation in Learning and Teaching Policy in the context of teaching and learning using AI systems.
    4. The University acknowledges that AI technologies and their associated governance requirements are rapidly evolving; therefore, this Policy, other university AI policies and frameworks as well as the roles and responsibilities of governance committees must remain flexible and agile, with monitoring, reporting, and review processes designed to adapt promptly to emerging risks, opportunities, regulatory and best practice changes.

  2. Policy principles for use of AI Systems

    1. AI Systems offer significant opportunities to enhance research, innovation, operational efficiency and service delivery. AI Systems also present risks to accuracy, impartiality, privacy, security and cultural appropriateness that must be carefully managed.
    2. AI Systems should be used to augment human capability, remain inclusive and accessible, and uphold the wellbeing and aspirations of students, staff and the wider community. Human judgement and responsibility must be retained for ultimate decision-making.
    3. AI System use by staff is expected to align with the University’s staff values:
      1. Manaakitaka | Respect: Use of AI Systems must protect personal information, honour intellectual-property rights, and uphold individual and collective authority over data, and respect Māori data sovereignty and Pacific data governance principles.
      2. Pono | Integrity: Use of AI Systems must be accountable, transparent, and explainable. Staff must ensure accuracy, disclose AI use appropriately, and maintain professional and ethical standards.
      3. Māhirahira | Curiosity: Staff should explore AI's potential to improve work processes while maintaining continuous learning about AI capabilities, limitations, and emerging risks.
      4. Whakawhanaukataka | Community: Use of AI Systems must be inclusive, accessible, and designed to deliver equitable outcomes without unfair discrimination or bias.
    4. AI System use must respect Aotearoa | New Zealand's unique cultural context, Te Ao Māori principles (rangatiratanga, kaitiakitanga, manaakitanga), and Pacific values of relationality and collective wellbeing.
    5. Staff, contractors and consultants are expected to consider the significant potential environmental and sustainability impacts of AI Systems, particularly those with significant computational requirements, and consider alternatives.

  3. Limitations on use of AI Systems

    1. The following uses of AI Systems are prohibited under this policy if undertaken without human oversight:
      1. processing content with the potential to harm the interests of individuals or communities, or
      2. producing content intended to represent the University’s position on any matter.
    2. AI System use must be disclosed where it significantly contributes to a work product, or where required by research ethics approvals, funding bodies or publishers, or where stakeholders would have a reasonable expectation of disclosure.
    3. AI use involving Māori data or Pacific data not otherwise approved under research ethics processes, or subject to appropriate consent, requires written endorsement from the Chief Digital Officer.
      1. Find further information on Pacific Data Sovereignty

  4. Procedures before using AI Systems

    1. Staff using AI Systems are expected to complete the University’s AI Literacy Training Induction, and consult relevant published materials, including:
      1. LibGuides: Generative AI
      2. Guidance For Classifying Information and Data at Otago
      3. Guidance For Data Storage Locations, which includes the AI approved tools list
      4. Use of Generative-Artificial Intelligences and Autonomous Content Generation in Learning and Teaching Policy
    2. Before deploying any new AI System, staff must notify the Cyber Security and Identity Office via cybersecurity@otago.ac.nz, so that an assessment of security and suitability may be undertaken, in accordance with the software approval process. Each AI System approved must have a designated AI System Owner accountable for that system’s lifecycle and compliance.
    3. Before processing any University-held data in any Generative AI System, staff must classify that data according to the Guidance For Classifying Information and Data. Only Generative AI Systems approved for the classification level of such data may be used.  The processing in a Generative AI System of information classified as ‘Private confidential’, ‘Business confidential’ or ‘Restricted’ requires the written approval of the Chief Digital Officer, which may be obtained via the Cyber Security and Identity Office.
    4. University information classified as ‘Internal Use’ may be processed in any AI Systems where staff can confirm using such systems would not breach classification guidelines. Where staff are unsure about the effects of using such systems staff can seek support from the Cyber Security and Identity Office cybersecurity@otago.ac.nz
    5. University information classified as ‘Public’ may be processed in any AI System.

  5. Procedures when using AI Systems

    1. When considering the output of an AI System, staff, contractors and consultants must recognise that AI System outputs can:
      1. reflect biases and gaps in their training data that may result in stereotypes, misrepresentation, or exclusion of diverse perspectives
      2. contain ‘hallucinations’, which are presented with confidence but contain entirely false information
      3. systematically misidentify or exclude perspectives from Māori, Pacific peoples, and/or other communities.
    2. Any use of an AI System’s outputs must retain human oversight, including:
      1. reviewing the output of an AI System in full before implementing decisions made by or based upon the output of an AI System
      2. maintaining human alternatives to any AI System processes and ensuring decision making processes remain explainable and auditable
      3. using academic and professional judgement to reject AI System outputs that may present risks to accuracy, impartiality, privacy or security, and
      4. cross-checking with other sources of information to ensure that the output is fair and unbiased, including appropriate consultation with cultural advisers.

Related policies, procedures and forms

Contact for further information

If you have any queries regarding the content of this policy, procedure, or guideline, or need further clarification, contact:

Chief Digital Officer
Email cdo@otago.ac.nz

Back to top