Session 3.04
Tracks
Track 4
| Thursday, June 18, 2026 |
| 1:55 PM - 2:55 PM |
| Sky Lounge |
Overview
Shadow AI use and behaviour change
Details
This session is suitable for: Intermediate
Speaker
Matt Jak
Director
Technology Systems Training
Shadow AI use and Behaviour change
Presentation description
Shadow AI in Schools: What It Is, Why It Matters, and What to Do About It
Session overview
78% of Australian professionals already use AI tools in their daily work. In schools, 38% of educators have shared sensitive student data with unvetted AI platforms. This session cuts through the noise to define shadow AI, show why it is a specific risk for Queensland schools, and give attendees five practical solutions they can take back and act on Monday morning.
Session structure
01 -- What is shadow AI? (8 min)
Defining the term. The difference between shadow IT and shadow AI. Why browser-based, free tools are invisible to IT teams and impossible to ban effectively.
02 -- Why schools are exposed (8 min)
Student data classifications, the Privacy Act 1988, and the Australian Framework for Generative AI in Schools. What Queensland government agencies are getting wrong that schools are repeating.
03 -- Real examples from Australian classrooms (8 min)
Case-based scenarios drawn from current EdTech and OAIC guidance. What happens when a teacher uploads a class list to ChatGPT. What the data actually does next.
04 -- Five solutions that work (14 min)
Audit and inventory. Governance frameworks. Sanctioned alternatives like NSWEduChat. Data loss prevention tools. AI literacy and culture change. Each addressed practically, not theoretically.
05 -- Q&A and action planning (7 min)
Attendees leave with one concrete next step mapped to their school context. Open discussion. Slide deck and reference list shared via QR code.
Key audience takeaway
Banning AI tools does not work. Providing safe, school-sanctioned alternatives combined with clear policy and staff training does. Attendees leave knowing the difference, and knowing what to ask their IT team on Monday
Session overview
78% of Australian professionals already use AI tools in their daily work. In schools, 38% of educators have shared sensitive student data with unvetted AI platforms. This session cuts through the noise to define shadow AI, show why it is a specific risk for Queensland schools, and give attendees five practical solutions they can take back and act on Monday morning.
Session structure
01 -- What is shadow AI? (8 min)
Defining the term. The difference between shadow IT and shadow AI. Why browser-based, free tools are invisible to IT teams and impossible to ban effectively.
02 -- Why schools are exposed (8 min)
Student data classifications, the Privacy Act 1988, and the Australian Framework for Generative AI in Schools. What Queensland government agencies are getting wrong that schools are repeating.
03 -- Real examples from Australian classrooms (8 min)
Case-based scenarios drawn from current EdTech and OAIC guidance. What happens when a teacher uploads a class list to ChatGPT. What the data actually does next.
04 -- Five solutions that work (14 min)
Audit and inventory. Governance frameworks. Sanctioned alternatives like NSWEduChat. Data loss prevention tools. AI literacy and culture change. Each addressed practically, not theoretically.
05 -- Q&A and action planning (7 min)
Attendees leave with one concrete next step mapped to their school context. Open discussion. Slide deck and reference list shared via QR code.
Key audience takeaway
Banning AI tools does not work. Providing safe, school-sanctioned alternatives combined with clear policy and staff training does. Attendees leave knowing the difference, and knowing what to ask their IT team on Monday
Biography
Matt Jak is the director of Technology Systems Training. He helps schools and purpose-led organisations use technology to reduce effort, so they can spend more time on the work that matters. Background in technology adoption and behaviour change across government and finance, sectors where AI getting it wrong has legal and human consequences. Matt's approach is evidence-led and values-anchored: AI works for teachers when it removes friction without creating risk, and when staff are actually trained, not just told. At DATTA QLD 2026, he is presenting two working AI tools built specifically for Queensland teachers, and running a separate practical workshop on shadow AI use and behaviour change.
