Australian school AI strategies are no longer a rarity.
Campion Education's EdTech Adoption in Australian Schools 2026 report shows that 61% of schools now have formal policies governing student use of generative AI.
As the national conversation shifts from whether to adopt AI to how to implement it, schools are increasingly seeking practical guidance on using it safely, consistently and in ways that genuinely lift teaching and learning.
At the recent CanvasCon ANZ in Sydney, Rachel Orston, Chief Customer Officer at Instructure and Professor Martin Bean CBE AM discussed the gap between having an AI strategy and putting it into practice.
Bean noted that almost every strategy commits to using AI ‘ethically and responsibly’, but this rarely tells teachers whether they can use it to draft a lesson plan, differentiate materials or prepare student feedback.
Orston said while considerable time has been spent telling teachers how careful they need to be with AI, far less time has been spent making clear what they’re actually allowed to do with it.
Governance should create permission
Orston said schools are only telling teachers half the AI governance story.
“We tend to think about AI governance as a set of restrictions: what information can’t be entered into a tool, which platforms can’t be used and which tasks shouldn’t be handed over to AI,” Orston told The Educator. “Those guardrails are essential.”
Good governance should also show teachers where the boundaries are and make clear that they're allowed to experiment within them, Orston added.
“When Martin and I discussed this, I suggested we could think of guardrails as permissions, a distinction that’s key,” she said, adding that when teachers don’t know where the boundaries lie, uncertainty can quickly become inaction or stall progress.
“Some will wait for further direction. Others will experiment but keep what they’re learning to themselves, unsure whether what they’re doing is school-sanctioned. Neither approach helps schools learn how to use AI effectively and responsibly.”
Orston said what became clear in her conversation with Professor Bean was that an AI strategy is most useful when teachers can see what responsible use looks like in practice.
“Can AI be used to develop ideas for a lesson? Can it help create differentiated classroom activities? Can teachers use it to help draft communications or feedback, provided they review and take responsibility for the final output? What information should never be entered into a public AI tool?” she said. “Teachers are already navigating these questions, and clearer guidance can make that easier.”
Permission needs to be supported by AI literacy
Orston said giving teachers permission to experiment keeps oversight in place and pairs clear boundaries with the knowledge teachers need to exercise professional judgment within them.
“This requires a shared level of AI literacy,” she said. “Teachers should understand how these tools work, their limitations and their risks well enough to make informed decisions about when they can support teaching and when human judgment must remain firmly in control.”
Orston said that includes understanding issues such as privacy and data security, recognising that AI-generated information can be inaccurate, knowing when disclosure of AI use is appropriate and considering what happens to learning when a task is handed over to a machine.
“Australia's Framework for Generative Artificial Intelligence in Schools provides an important starting point, meaning schools don't need to develop their approach to AI from scratch,” she said. “The opportunity now is to translate these national principles into practical guidance and a shared understanding of what AI can do, where human oversight is essential, how student data should be protected and what successful learning outcomes look like.”
Orston said this common understanding can help school leaders and teachers make more informed decisions about AI tools, while giving technology providers clearer expectations about how their products should support educational goals.
Students are already setting their own boundaries
One of the most interesting lessons around AI use came from students themselves, Orston said.
“A panel of learners at different stages of their education shared how they decide when to use AI and when not to,” she said. “One student described using AI to write notes, before realising that the writing process was part of how he actually learned the material.”
Orston said that by automating the task, the student also inadvertently removed part of his learning.
“A secondary student raised another issue schools should pay close attention to: transparency. When teachers use AI without acknowledging it, students can begin questioning whether they can trust the material they're being taught, and the person teaching it,” she said. “There’s an important lesson in both observations.”
Orston said using AI well means understanding what a tool can do and what we might lose when we hand a task to it.
“Students are already wrestling with those questions,” she said. “Their comments also highlighted the importance of reciprocity: if students are asked to use AI transparently, they'll likely expect the same from educators.”
Let classroom experience shape the strategy
While AI is quickly becoming a fixture of the modern classroom, there is no settled playbook for AI in education yet, Orston cautioned.
“Schools shouldn't pretend there is,” she said. “Strategies need to evolve as teachers and students discover what works, what doesn't and where unexpected problems emerge.”
One takeaway, said Orston, is that strategies are more likely to evolve when teachers have room to put policy into practice and share what they learn.
“Staff forums, professional learning sessions and regular check-ins are some of the ways schools can create space for those lessons to be shared,” she said. “That feedback can also give school leaders a clearer view of where guidance may need to evolve.”
For school leaders, Orston said she would start with three questions: do our teachers know what they're allowed to try with AI?; do they have somewhere to share what they learn when they try it?; when did we last change our AI guidance based on the experience of teachers and students?
“Australian schools have put considerable work into developing AI strategies. The next challenge is making sure those strategies give teachers the confidence, knowledge and permission to act.”