Quick Answer: Lane 1 assessments are controlled tasks — exams, invigilated tests, oral assessments — where AI use is restricted by default. Lane 2 assessments are uncontrolled tasks — essays, reports, case studies, portfolios — where AI use is generally permitted but must follow course-specific disclosure rules. The University of Auckland's Senate endorsed this Two-Lane Approach in November 2025, the terminology is becoming visible to students through 2026, and it will apply across every course university-wide by 2027.
If you've submitted an assignment recently and genuinely didn't know whether using ChatGPT for a first draft would get you flagged, you're not overthinking it — the rules really have been inconsistent between courses. New Zealand universities are now trying to fix that ambiguity properly, and the "Lane 1 / Lane 2" system is the clearest answer to come out of it so far.
What Is the Two-Lane Approach — And Why Now?
The Two-Lane Approach was adopted at the University of Auckland to give teaching staff a consistent way to design assessments around the fact that generative AI tools are now genuinely ubiquitous among students. Rather than leaving each lecturer to write their own ad-hoc AI policy into a course outline, the university now requires every assessment to be explicitly labelled as either Lane 1 or Lane 2.
The intent isn't to police AI use for its own sake — it's what the university describes as "intentionality." Some assessments exist specifically to verify what you personally know and can do without assistance. Others exist to build the kind of applied, AI-integrated skills you'll actually use in a workplace after graduation. The two-lane label simply makes it explicit, up front, which kind of task you're dealing with.
Many courses already started using this language through 2026. Full, university-wide rollout is scheduled for 2027, so if your course hasn't labelled its assessments yet, it likely will soon.
Lane 1 Explained: What Counts, What's Restricted
Lane 1 covers controlled assessment — situations where the university needs a trustworthy, individual picture of what you actually know, without AI in the mix unless your course specifically says otherwise. This typically includes:
- Invigilated, in-person exams
- Interactive oral assessments and vivas
- Timed, supervised tests
- Practical work assessed under direct observation
The purpose of Lane 1 is assurance — it's the university's way of confirming that the person who submitted the essay in Lane 2 is the same person who actually understands the material, without needing to distrust every piece of work you hand in.
Lane 2 Explained: What "AI Allowed" Actually Means
Lane 2 covers uncontrolled assessment — work completed outside supervised, invigilated conditions. This includes essays, portfolios, case studies, written reports, presentations, and most take-home coursework.
Here, AI use is not restricted by default, but that doesn't mean it's a free-for-all. Courses running Lane 2 assessments commonly still expect you to:
- Disclose how and where you used AI tools in your process
- Keep a log of prompts or drafts if your course requires it
- Critically evaluate and correct anything AI produced, rather than submitting it unedited
- Meet the same standard of originality, referencing, and rubric alignment as any other submission
This is the part students most often get wrong.
The Common Mistake: "Lane 2 ≠ Anything Goes"
The single clarification the University of Auckland makes most explicitly is this: don't assume Lane 1 means "no AI" and Lane 2 means "anything goes." Lane 2 still comes with conditions — disclosure expectations, process documentation, and the underlying requirement that submitted work reflects your own understanding. A Lane 2 label is permission to use AI as a tool in your process, not permission to submit AI output as your final work.
This lines up with NZQA's broader academic integrity guidance, which continues to require that any generative AI use be transparent and that institutions regularly review their policies as the technology changes.
Rollout Timeline: 2026 vs 2027
| Year | What Happens |
|---|---|
| 2025 (Nov) | University of Auckland Senate formally endorses the Two-Lane Approach |
| 2026 | Many individual courses begin labelling assessments as Lane 1 or Lane 2; students should check with their Course Coordinator if it isn't yet stated |
| 2027 | Every course and programme at the university implements the Two-Lane Approach in full |
If your course guide or assignment brief doesn't mention a lane yet, it doesn't mean AI use is automatically fine — it means the default course rules on academic integrity still apply, and it's worth confirming directly with your lecturer before you use AI on anything.
How This Compares Across Universities
Auckland isn't working in isolation here. The University of Sydney runs a near-identical two-lane framework, where Lane 1 is "secure" assessment conducted under supervision and Lane 2 is "open" assessment where AI use cannot reasonably be banned or controlled. Sydney's version has been rolling out since 2024, giving Auckland a tested model to build from rather than starting from scratch. Expect other NZ institutions to move toward similar language over the next few years, even if they don't use the exact same "lane" terminology yet.
What This Means for Your Assignments, Practically
- Check your course outline first. If it explicitly says Lane 1 or Lane 2, follow that label's rules exactly.
- If nothing is labelled, ask. Don't assume — email your Course Coordinator rather than guessing.
- In Lane 2 work, document your process. Keep drafts and, if asked, a simple log of where you used AI assistance.
- Never submit AI output as final work, even in Lane 2 — it still needs to be your own critically reviewed writing, correctly referenced to your faculty's required style.
- Treat AI-generated material the way you'd treat a source — something to learn from and build on, not something to hand in directly.
This is exactly the same standard we cover in more depth in our guide on using ChatGPT for NZ university assignments under the 2026 rules, and in our breakdown of whether Turnitin's AI detector actually works if you're worried about false flags on genuinely original writing.
FAQ
Q. What is the Two-Lane Approach to assessment?
It's a University of Auckland framework that labels every assessment as either Lane 1 (controlled, AI-restricted) or Lane 2 (uncontrolled, AI-permitted with conditions), so students know exactly what's allowed on each task.
Q. Does Lane 2 mean AI use is unlimited?
No. Lane 2 permits AI as part of your process, but you're still expected to disclose your use where required, critically edit any AI output, and submit work that reflects your own understanding.
Q. Which New Zealand universities use Lane 1 and Lane 2?
The University of Auckland formally adopted this model in November 2025, with rollout through 2026 and full implementation by 2027. Similar two-lane frameworks are already used internationally, including at the University of Sydney.
Q. When does the two-lane system become mandatory at Auckland?
Full, university-wide implementation is scheduled for 2027, though many individual courses started using the terminology during 2026.
Q. Do I need to disclose AI use in Lane 2 assessments?
Often yes — many Lane 2 tasks require you to declare how you used AI tools or keep a record of your process. Check your specific course guide, since requirements vary by course.