Let’s be honest. You have probably already used AI. Maybe to write an email. Fix a lesson plan. Or just answer a quick question.
And it saved you time. A lot of time.
But here is the thing nobody tells you — one wrong AI prompt can expose your students’ private data, invalidate your assessments, and land your RTO in front of ASQA.
Not because you did something terrible. Because nobody told you the rules.
This guide fixes that.
It is written for RTO managers, trainers, admin teams, and compliance officers who want to use AI properly — and stay ahead of every audit.
The message is clear. AI is everywhere. The rules are real. And ASQA is watching.
Before worrying about what not to do — understand what AI is genuinely good for. Used right, AI is one of the most powerful tools your RTO has ever had.
| Area | What AI Can Do | Time Saved |
| Admin | Draft generic email templates | 1–2 hrs/day |
| Trainers | Generate lesson plan ideas | 1 hr/session |
| Compliance | Summarise policies in plain English | 30 min/doc |
| Marketing | Rewrite brochures and course descriptions | 50% faster |
| Student Support | Create response templates for common requests | 1 hr/day |
Practical Example:
A trainer preparing a BSB30120 unit used to spend 3 hours writing activity ideas from scratch. With AI, they typed: “Give me 5 practical team-building activities for a Certificate III Business class.” Got 8 ideas in 10 seconds. Chose 3. Adapted them. Done in 20 minutes.
That is not cheating. That is smart.
The key rule: AI assists your team. It does not replace their judgement, qualifications, or professional responsibility.
This is the most important step. No exceptions.
The Privacy Act 1988 and Australian Privacy Principles (APPs) apply to every AI tool that touches personal information — including public tools like ChatGPT.
When you paste student or staff information into a public AI tool, that data may be stored, processed offshore, and used to train future AI models. You lose control of it permanently.
Up to $50 million per breach — or 30% of annual turnover, whichever is greater. From June 2025, individuals can also sue your RTO directly without proving financial damage.
| DANGEROUS | SAFE |
| “Fix assessment for John Smith, ID 12345, CHC50121” | “Fix generic childcare assessment example” |
| “Email to student Sarah about her transfer” | “Create student transfer request email template” |
| “Summarise trainer James’s performance review” | “Summarise generic trainer performance review” |
| “Help with complaint from student in Melbourne” | “Create student complaint response template” |
Practical Example:
Student Services needed to respond to 20 deferral requests. Instead of typing each student’s name into AI, they used: “Write a professional, empathetic email template for student deferral requests.” Copied the template. Filled in the student name manually. 20 emails done in 40 minutes. Zero privacy risk.
ASQA’s 2026 workshops make this clear: AI compliance starts with your people, not your tools.
An AI policy on paper means nothing if your team does not know it exists.
| Role | What They Need to Know |
| CEO/Compliance Manager | Policy, legal obligations, audit documentation |
| Trainers | What students can/cannot use AI for, detection tools |
| Admin/Student Support | Safe prompts, approved tools, privacy rules |
| Marketing | Generic prompts only, no student references |
| Finance | No financial or personal data in AI tools |
Run this with your team this week:
Practical Example:
One Melbourne RTO trained their team in 30 minutes before April 1. Result: Zero AI-related incidents in their next ASQA review. The audit specifically noted their “documented AI governance process” as a strength.
Ongoing Training: Repeat every 12 months or when a new AI tool is introduced.
Your team needs clear rules — not grey areas. Write a one-page AI policy and display it in the office.
Reviewed annually by: Compliance Manager
Be specific in your Student Handbook:
| AI Task | Student Allowed? |
| Use AI for research ideas | Yes (disclose) |
| Use AI to improve grammar | Yes (disclose) |
| Use AI to write the full assessment | No |
| Use AI to generate evidence of competency | No |
| Submit AI output as their own work | No |
ASQA’s position (2026): ”AI cannot make assessment decisions. Human must always validate.” Co-assessment with AI is not a valid assessment method.
ASQA confirms academic integrity — including AI cheating — is a compliance risk they monitor actively.
Watch for:
Free Detection Tools
Practical Example:
Trainer receives a polished report from a student who usually struggles. Runs ZeroGPT — 92% AI generated. Asks the student: “Can you walk me through how you answered question 3?” Student cannot explain. Trainer asks for resubmission with a verbal component. Competency properly demonstrated. Assessment valid.
AI USE LOG
Date: __/__/____
Staff Name: ________________
Task: ____________________
Tool Used: ________________
Input Used (generic only): ________________
Output Reviewed By: ________________
Human Decision Made: Yes / No
Final Document Approved By: ________________
AI tools change fast. Your policy needs to keep up.
Set a calendar reminder every quarter: “AI review — 15 minutes.” Check with your compliance manager: any new tools? Any incidents? Any ASQA updates? Update the log. Done.
ASQA does not just want a policy — they want evidence you are actively managing AI risk. A quarterly review log shows exactly that.
Yes — for research ideas, brainstorming, and grammar improvement, if they disclose it. No — for writing full assessments, generating competency evidence, or submitting AI output as their own work.
The free version — no. It stores your input and uses it to train future models. Use Microsoft Copilot Enterprise or Grammarly Business, which have privacy protections for business use.
No. ASQA is clear — humans must make all assessment decisions. AI can help you draft feedback ideas, but the trainer must read, validate, and sign off.
Use a verbal discussion component. Ask the student to explain their work. Genuine competency shows up in conversation. AI-assisted work often does not.
ASQA has not yet made an AI policy a mandatory document. But their 2026 workshops strongly imply it is expected — and it protects your RTO if an issue arises.
No. Never. These involve sensitive personal information and require human judgement, empathy, and legal compliance. Use AI only for generic template ideas — never input real case details.
Up to $50 million per serious breach — or 30% of annual turnover, whichever is greater. From June 2025, individuals can also sue your RTO directly in court.
Check the tool’s privacy policy. Key questions: Does it train on your data? Is data stored outside Australia? Enterprise-licensed tools like MS Copilot and Grammarly Business have contractual data protections.
Yes — for generic tasks like summarising legislation, drafting templates, or suggesting process improvements. Never input actual student data, audit evidence, or staff information.
Minimum once a year — or whenever a new tool is introduced or ASQA releases new guidance.
AI is not going away. RTOs that use it safely will save time, improve quality, and impress auditors. RTOs that ignore it will fall behind. RTOs that use it recklessly will face privacy fines, invalid assessments, and audit findings.
The choice is yours. The rules are clear. And now you have the roadmap.
Disclaimer:
The information presented on the VET Resources blog is for general guidance only. While we strive for accuracy, we cannot guarantee the completeness or timeliness of the information. VET Resources is not responsible for any errors or omissions, or for the results obtained from the use of this information. Always consult a professional for advice tailored to your circumstances.
Ben Thakkar is a Compliance, Training, and Business specialist in the education industry. He has held senior management roles, including General Manager, with leading Registered Training Organisations (RTOs) and Universities. With over 15 years of experience, Ben brings extensive expertise across audits, funding contracts, VET Student Loans, CRICOS, and the Standards for RTOs 2025.
By submitting this form, you agree to the VET Resources Privacy Policy.
