The state of AI adoption at the University of Sydney

The Sydney Assessment Framework was first published as a two-lane FAQ on 2 July 2024 and consolidated as a formal explainer in March 2025. Lane 1 is 'secure' assessment of learning, conducted face-to-face under supervision. Lane 2 is 'open' assessment for and as learning, where, per the FAQ, 'the use of generative AI tools cannot be banned, restricted, or controlled'. From Semester 1 2025 the default changed to allow AI in assessments except where staff opt out; Semester 2 2025 rolled out the full two-lane model; assessment plans for every degree are due by end of 2025.^[1][2]

On AI detection tools, Sydney uses Turnitin's AI detector on a discretionary basis when AI use is suspected, but explicitly does not treat the score as standalone evidence: 'the AI detector score would not be the only evidence relied upon for an academic integrity case, but will be considered alongside other relevant evidence.'^[4] The two-lane FAQ goes further: 'It is already not possible to restrict or detect AI use in assessments which are not secured by face-to-face supervision', framing the two-lane model itself as 'the only reliable, tenable, and equitable way to protect the integrity of our degrees.'^[2]

Pro Vice-Chancellor (Educational Innovation) Adam Bridgeman has stated the institution's pragmatic position directly: 'It's no secret that any take-home assessment can be completed to a high level by AI, and it's not practical, enforceable or desirable to implement a blanket ban on these tools.'^[3] Deputy Vice-Chancellor (Education) Professor Joanne Wright has framed the workforce rationale: 'Generative AI has already had a profound impact on workplaces and our graduates are expected to demonstrate skilled use of the relevant tools in job interviews.'^[3]

Bloom runs on Microsoft Azure OpenAI under enterprise data handling, the same model infrastructure Sydney uses for Cogniti within its institutional tenancy. Prompts and responses are not used to train base models, and Bloom's data posture sits inside the same vendor relationship Sydney's Educational Innovation team already operates against.^[5]

Bloom can be deployed at Sydney in a day, with course-specific material ingestion and convenor-controlled student access. No new procurement, no new vendor security review, and no engineering integration is required.

Sydney's tooling layer pairs a vendor product (Microsoft Copilot for Web) with an in-house build (Cogniti).

Sydney's institution-level AI research is anchored by the Sydney Artificial Intelligence Centre, sitting within the Faculty of Engineering and directed by Dr Tongliang Liu. The Centre's stated focus is fundamental AI (algorithms, learning theory, transfer learning with causal structures) plus applied work in computer vision, multimedia and cybersecurity.^[6]

On AI governance, the Centre for AI, Trust and Governance (CAITG) launched on 25 February 2025 within the Faculty of Arts and Social Sciences. CAITG is co-directed by Professor Terry Flew (Digital Communication and Culture; ARC Laureate Fellow) and Professor Kimberlee Weatherall (Sydney Law School), focusing on the institutional, legal and policy dimensions of AI rather than the technical research stack.^[7]