Every Australian university has an AI policy now. The problem is that most of them read like they were written by a compliance team for another compliance team, then buried in a PDF nobody bookmarks. If you’re a student sitting down tonight to write an assignment, and your only question is “can I use ChatGPT for this?”, you’re going to have a frustrating time finding a straight answer.

I’ve been through this myself. When I did my MBA at Swinburne (2017 to 2020), generative AI wasn’t even part of the conversation. Now I’m studying IT at QUT while working full-time, and the rules around AI use change depending on which subject I’m in, which assessment I’m doing, and sometimes which week of the semester it is. It took me longer than I’d like to admit to figure out how the system actually works.

This article translates the regulatory language and institutional policies into something you can act on. I’ll walk you through how the system is structured, why there’s so much inconsistency, and give you a practical framework for working out what’s allowed before you start any assessment.

Australia Doesn’t Have One AI Rule. It Has Hundreds.

The first thing that surprises most students is that there’s no single national AI policy for universities. The Tertiary Education Quality and Standards Agency (TEQSA), Australia’s higher education regulator, sets the broad framework. Their August 2024 guidance paper describes generative AI as an “evolving risk to academic integrity” and outlines a range of responses institutions can take (TEQSA, 2024).

But TEQSA doesn’t tell your university exactly what to allow or ban. Each institution writes its own policy. The University of Sydney has a detailed student-facing page. The ANU publishes a best-practice guide with specific do’s and don’ts. Smaller providers like the Australian Institute of Higher Education and the Australian Institute of Business each have their own frameworks that differ again (AIH, 2024; AIB, 2024).

What this means in practice: the rules your friend follows at Deakin may be completely different from what applies to you at UQ or Macquarie. You can’t copy someone else’s approach. You need to read your own university’s policy.

The Spectrum: Prohibited, Permitted, and Everything in Between

Most Australian universities now sit somewhere on a spectrum. At one end, certain assessment tasks prohibit AI entirely. This is common for exams, in-class tests, and assessments specifically designed to evaluate your unassisted knowledge. At the other end, some units actively encourage or even require AI use, asking students to critique AI output or demonstrate AI-augmented workflows.

The critical distinction is in the middle. Using AI as a thinking aid (brainstorming, summarising your own notes, checking your understanding of a concept) is treated very differently from submitting AI-generated text as your own work. The first is generally accepted at most institutions. The second is treated as academic misconduct, and at many universities it’s classified alongside contract cheating.

TEQSA’s student advice page makes this explicit: “there might be subjects or tasks where the use of AI is encouraged or even required,” but using AI “inconsistently with institutional rules” constitutes misconduct (TEQSA, 2023). The same regulator in the same document manages to say both “you might need AI” and “using it wrong can end your degree.” Context is everything.

And the context shifts not just between universities, but within them. Your Faculty of Business might take a different stance from your Faculty of Arts. Your first-year foundational subject might ban AI completely while your capstone subject requires you to use it. TEQSA acknowledges this directly: students should “make sure you understand the expectations for each assessment task,” not just each course (TEQSA, 2023).

The Two-Step Check: How to Know What’s Allowed

Here’s the practical framework I use before starting any assessment.

Step one: read the university-level AI policy. This sets the default position. It tells you your university’s general stance on AI, what the misconduct definitions look like, and what the baseline disclosure requirements are. Most universities publish this on their academic integrity page. The University of Sydney’s AI page is a solid example of what to look for.

Step two: read the unit outline for that specific assessment. This is the step most students skip, and it’s the one that matters most. The unit outline (sometimes called a course profile or subject guide) specifies whether AI is permitted, restricted, or prohibited for each individual assessment task. If the unit outline says something different from the university-level policy, the unit outline governs for that task.

If neither document says anything about AI, treat it as prohibited until you’ve confirmed otherwise. Send your unit coordinator an email asking for clarification. It takes two minutes, and it’s the cheapest insurance you’ll ever buy.

I keep a simple note for each subject at the start of semester: the AI policy position and any assessment-specific exceptions. It saves me from re-checking every time I sit down to work. For a more detailed look at how to decode what your assessments are actually asking for, I’ve written about how to read a university rubric as a skill in its own right.

Why Mature-Age Students Carry More Risk

If you’re coming back to study after time away from university, the AI policy landscape is genuinely harder for you than it is for a school-leaver who’s been in the system the whole time.

When I started my MBA in 2017, academic integrity meant “don’t copy someone’s essay” and “use Turnitin before you submit.” There was no ChatGPT. There were no AI disclosure requirements. There was no AI detection layer in Turnitin. The framework has shifted fundamentally since then, and the transition happened while most mature-age students were out of the system, working, raising families, living their lives.

You come back and three things compound at once. First, the institutional guidance is scattered across policy PDFs, unit outlines, library guides, and (sometimes contradictory) verbal advice from teaching staff. Second, the rules are genuinely complicated; even current students find them confusing. Third, you don’t have the benefit of corridor conversations and peer osmosis that full-time on-campus students get.

During the research phase where I was validating whether GradeMap should exist as a real product, I spoke with mature-age students who described the same pattern. One student, a single mum juggling work and study, said she “never, ever relied on anyone” for help. Another, studying online part-time, only had 30 to 60 minutes between commitments. When your study happens in stolen pockets of time, you can’t afford to spend half that session hunting down which AI rules apply to tonight’s assignment.

TEQSA has recognised this gap to some extent. Their student resources hub catalogues AI literacy modules from universities across Australia, including CQUniversity’s generative AI literacy module and Flinders University’s “Using AI tools for study” guide. They’ve also translated their student advice into multiple languages, signalling that international students face similar navigational challenges (TEQSA, n.d.).

If you’re feeling behind on all of this, know that the two-step check above puts you ahead of most students who’ve never read the policy at all. And if you’re a working parent studying in 30-minute blocks between pick-up and bedtime, the last thing you need is to lose marks (or worse) because of a rule you didn’t know existed.

Disclosure and Citation: The Messy Reality

Even when AI use is explicitly permitted, you still need to disclose it. But how you disclose it varies wildly.

Some universities require an AI use statement on every submission, whether you used AI or not. Others only want disclosure when AI contributed substantively. Some expect a detailed log: the tool you used, every prompt you entered, and how the output shaped your work. Others are satisfied with a brief acknowledgment.

The referencing side is just as inconsistent. UNSW’s referencing guide gives one format. UQ’s library guide gives another. Bond University has its own approach again. They’re all technically “APA” or “Harvard” format, but the specifics differ enough to trip you up if you’re following the wrong university’s template.

My advice: check your unit outline first for any assessment-specific disclosure requirements. Then check your university’s library guide for the correct citation format. When in doubt, over-disclose. No student has ever been penalised for being too transparent about their process. If you want a detailed walkthrough of the mechanics, I’ve covered how to cite AI in your university assignments in a separate guide.

AI Detection Tools: Useful but Flawed

Many Australian universities use Turnitin’s AI detection feature, and you should understand what it can and can’t do.

Turnitin’s detector analyses writing patterns and produces a probability score estimating whether text was AI-generated. It’s not a binary verdict. Universities set their own thresholds for what score triggers further investigation, and most treat the score as a starting point, not proof.

The problem is false positives. A 2023 study published in Patterns found that AI detectors are significantly biased against non-native English speakers, flagging their writing as AI-generated at substantially higher rates than native speakers’ writing (Liang et al., 2023). Stanford’s Human-Centred Artificial Intelligence institute covered this finding in detail, noting the real-world consequences for students who write in formal or precise styles (Stanford HAI, 2023).

This isn’t theoretical. An ABC News investigation documented cases where Australian Catholic University wrongly accused students of AI cheating based on detection tool results (Bergin, 2025). Students who had written entirely original work were flagged and forced through misconduct processes.

What does this mean practically? Keep your drafts. Save your research notes. If you use a Google Doc, the version history becomes evidence of your writing process. If you hand-write notes before typing, photograph them. This documentation protects you whether or not you used AI. And if you did use AI in a way that was permitted, your disclosure statement is your first line of defence.

For a deeper look at where detection technology is heading, I’ve written about AI detection tools and their trajectory in Australian universities.

Assessment Redesign: Why Your Exams Might Look Different

TEQSA isn’t just telling universities to catch cheaters. They’re actively encouraging institutions to redesign assessments so that AI misuse becomes a smaller problem in the first place.

Their assessment reform guidance recommends a shift away from detection-based policing toward assessment design that demands genuine engagement: critical thinking, personal application, and real-world problem-solving (TEQSA, n.d.). In practice, that means more oral presentations, more reflective portfolios, more assessments tied to lived experience, and fewer take-home essays where the gap between “student wrote this” and “AI wrote this” is hard to detect.

TEQSA’s student advice page endorses this directly: “Redesigning assessments is an appropriate response to the risks posed by AI and is one that TEQSA supports” (TEQSA, 2023). In June 2024, TEQSA issued a request for information to every Australian higher education provider asking for credible AI action plans. Every single provider responded (TEQSA, 2024). That 100% response rate tells you this isn’t optional.

Universities like UQ have published rubric principles specifically designed for assessments where AI is permitted, and Southern Cross University developed a seven-component lifecycle model for adapting assessments to the AI era. If your assessments feel different from what you expected, or if your unit coordinator has changed the format mid-semester, this is likely why. It’s a sector-wide shift, not an arbitrary decision by one lecturer.

TEQSA also has a specific note for students affected by these changes: if assessment redesign impacts you due to disability or other circumstances, you should raise this with your institution. All universities are required to have complaint and appeal procedures under the Higher Education Standards Framework (TEQSA, 2023).

A Practical Summary You Can Use Tonight

Here’s the short version for when you’re about to start an assignment.

Generally fine (but always check your unit outline):

Almost always requires disclosure:

Almost always prohibited:

Your checklist before every assessment:

  1. Read your university’s AI policy page
  2. Read the unit outline for that specific assessment
  3. If unclear, email your unit coordinator
  4. Document your process as you work (drafts, notes, version history)
  5. Disclose any AI use, even when you think it’s obviously allowed

Where GradeMap Fits

I’m building GradeMap because navigating AI policy is fundamentally a rubric literacy problem. You need to read and interpret official documents (your policy, your unit outline, your rubric), extract the rules that apply to your specific task, and apply those rules under time pressure. That’s the exact problem GradeMap is designed to help with: turning structured institutional language into a clear action plan you can follow.

For a returning mature-age student who didn’t grow up with academic integrity frameworks built around AI, that translation layer is especially useful. I know because I needed it myself.

References

Australian Institute of Business. (2024). Academic integrity policy. https://www.aib.edu.au/wp-content/uploads/2024/03/Academic-Integrity-Policy.pdf

Australian Institute of Higher Education. (2024). Artificial intelligence (AI) policy. https://aih.edu.au/wp-content/uploads/2024/06/ACAHE33-Artificial-Intelligence-Policy-2024.1.pdf

Bergin, J. (2025, October 9). University wrongly accuses students of using artificial intelligence to cheat. ABC News. https://www.abc.net.au/news/2025-10-09/artificial-intelligence-cheating-australian-catholic-university/105863524

Liang, W., Yuksekgonul, M., Mao, Y., Wu, E., & Zou, J. (2023). GPT detectors are biased against non-native English writers. Patterns, 4(7), 100779. https://www.sciencedirect.com/science/article/pii/S2666389923001307

Stanford Institute for Human-Centred Artificial Intelligence. (2023). AI-detectors biased against non-native English writers. https://hai.stanford.edu/news/ai-detectors-biased-against-non-native-english-writers

TEQSA. (2023). Artificial intelligence: Advice for students. https://www.teqsa.gov.au/students/artificial-intelligence-advice-students

TEQSA. (2024). The evolving risk to academic integrity posed by generative artificial intelligence: Options for immediate action. https://www.teqsa.gov.au/sites/default/files/2024-08/evolving-risk-to-academic-integrity-posed-by-generative-artificial-intelligence.pdf

TEQSA. (2024). Gen AI strategies for Australian higher education: Emerging practice. https://www.teqsa.gov.au/guides-resources/resources/corporate-publications/gen-ai-strategies-australian-higher-education-emerging-practice

TEQSA. (n.d.). Gen AI: Academic integrity and assessment reform. https://www.teqsa.gov.au/guides-resources/higher-education-good-practice-hub/gen-ai-knowledge-hub/gen-ai-academic-integrity-and-assessment-reform

TEQSA. (n.d.). Gen AI: Student resources and support. https://www.teqsa.gov.au/guides-resources/higher-education-good-practice-hub/gen-ai-knowledge-hub/gen-ai-student-resources-and-support

University of Sydney. (n.d.). Artificial intelligence. https://www.sydney.edu.au/students/academic-integrity/artificial-intelligence.html

FAQ

Can I use ChatGPT or Claude for my university assignments in Australia?

It depends on your university’s policy and the specific assessment task. Most Australian universities allow AI for learning activities like brainstorming and self-testing, but prohibit submitting AI-generated text as your own work. The rules often differ between assessments within the same subject. Always check your university’s AI policy page and your unit outline before using any AI tool on assessed work. If neither document mentions AI, treat it as prohibited until you’ve confirmed with your unit coordinator.

What happens if I’m accused of using AI when I didn’t?

False positives from AI detection tools are a documented problem, particularly for non-native English speakers and students with formal writing styles. If you’re accused, your best defence is evidence of your writing process: drafts, notes, version history, research trails. Every Australian university is required by TEQSA to have formal complaint and appeal procedures. Ask for the specific evidence being used against you, and request a meeting where you can explain your work and demonstrate your understanding of the material.

Do I need to declare AI use even if my university allows it?

Almost certainly yes. Most Australian universities that permit AI use require some form of disclosure. The specifics vary: some want a statement on every submission, others only when AI contributed substantively. Check your unit outline for assessment-specific requirements, and your university’s library guide for the correct citation format. When in doubt, over-disclose. Transparency is always safer than silence.

Why do AI policies differ so much between universities?

TEQSA sets the regulatory framework but doesn’t prescribe specific AI rules. Each university, faculty, and sometimes individual unit coordinator decides how to implement the policy for their context. A computing subject might encourage AI as a professional tool, while a philosophy subject might ban it to assess independent reasoning. This patchwork is intentional, reflecting the view that different disciplines and assessment types warrant different approaches to AI.

How do I keep up as policies keep changing?

Check your university’s academic integrity page at the start of each semester, and re-read each unit outline before starting every assessment. Policies are evolving rapidly as institutions respond to new TEQSA guidance and learn what works. What was banned last semester might be permitted this semester, and vice versa. Making the two-step check (university policy, then unit outline) a habit is the simplest way to stay current without drowning in policy documents.