Principal

With interest I recently read Pope Leo’s encyclical about AI, and its potential impact on our students, staff and community. In light of this, I have written my newsletter article using Co-Pilot – our Catholic Education approved AI tool – using numerous prompts to produce the final draft. I can tell you it is no easy task to do – our tendency is to allow automation to make things easier for us – a distinct trap when it comes to thinking. It is also very, very difficult for students to do so when they have not been formed in their thinking to such a level that they can make rational assessments about the inferences that AI makes about topics, knowledge, and/or humankind. Please read the following section knowing that even though AI mostly wrote the article, it does reflect my thinking about AI and student learning. In addition, I have read the texts cited in the reference list below and endorse only the specific claims I (via Co-Pilot) have cited in-text.

---------------------------------------------------------

Magnifica Humanitas

Pope Leo XIV’s Magnifica Humanitas sets a clear position. Technology must serve the human person. It must not replace the intellectual and moral work required of the learner or the teacher. Education is not the production of answers - that is the purpose of AI - but the formation of the person capable of judging, thinking and acting in truth. Catholic schools teach students, but they also form them morally and spiritually. Pope Leo’s encyclical arrives to us at a pivotal time in the age of technology. Society is accelerating at an astonishing pace toward development of complex AI systems, that to date, have no regulation other than that imposed by the programmer or the company, and further, we have no real idea about AI’s impact on our children.

AI’s intersection with our society is none more telling than in our schools. Here we hold many questions about the use of AI by students, and by broader implication by all people. AI will either improve learning or diminish it. There is no neutral position as far as I can see. Recent Australian research reinforces this. Artificial intelligence, cognitive offloading and implications for education (Lodge & Loble, 2026) identifies the central risk – which is not plagiarism. The central risk is that students begin to outsource the thinking required to learn – a term coined as cognitive unloading. When this occurs, learning becomes task completion rather than knowledge acquisition and reasoning. The article puts it succinctly:

“The true educational risk of AI is not simply that students will use it to cheat on an essay. The far more profound risk is that AI may fundamentally interfere with the cognitive processes of knowledge construction and verification, the very processes that build the long term memory stores and subsequent skills upon which the majority of critical thinking depends (Lodge & Loble, 2026. P12).”

The same report describes what it terms a “performance paradox”. Students can produce better immediate work using AI, but their learning declines once the support is removed. In other words, performance improves, learning does not. There is an inverse relationship between the quality of work student produced and the level at which they can analyse, abstract, deduct, or synthesise. Our work as educators is not only for the students to produce quality assessment, but our work is also measured by what and how our students can think, know and do independently. The issue is not that AI systems make decisions for students, but that they make it easy for students to avoid making decisions themselves. When students outsource their cognitive effort to AI—which are neural networks trained on large datasets—they risk bypassing the very thinking processes required to develop their independence. Independence is not gained through access to answers, but through the disciplined practice of judgment, evaluation, and problem-solving. If AI replaces that process, independence declines. If it is used to prompt and extend that process, independence may be strengthened.

Anna Goldsworthy (2026, The children section) in The God we Made makes a coherent point:

“Although our current LLMs (large language models) are only fledgling gods, they are already fluent in many of our love languages: corporate jargon, therapy, condolence card. If they have “knowledge,” it is a knowledge acquired differently to our own, inferred from symbols, rather than harvested empirically from the world. Until now, it has been up to humans to provide the sensory input: to be the seeing and hearing and smelling and feeling nodes for these giant pattern-seeking machines. They translate this input into tokens, which they arrange probabilistically to generate output that looks like thought.”

In this observation, Goldsworthy reinforces a critical distinction for schools. What appears as “thinking” in AI is, in fact, the rapid reassembly of patterns drawn from vast datasets. It is inference, not understanding; probability, not judgement. Unlike the human person —who knows through experience, reason, and moral reflection—AI produces outputs detached from lived reality and responsibility. This matters deeply in schools. If students begin to treat AI-generated responses as equivalent to thinking, they risk confusing fluency with truth and speed with wisdom. The task of education, therefore, remains unchanged: to form young people who can discern, evaluate and judge, not merely generate responses. AI may assist in presenting ideas, but it cannot replace the human act of coming to know, which is always grounded in effort, formation and discernment of truth.

Cairns Diocese Generative AI Position Statement

The Catholic Education Diocese of Cairns (2025) generative AI position statement reinforces these points. It is explicit that generative AI should enhance and must not diminish or replace the essential teacher–student relationship. It also requires that AI use preserves human control, professional judgement and accountability, and that student use must support learning rather than replace learning. This is why our approach is deliberate. Learning and thinking is not only about producing an answer. It is about engaging in the cognitive processes that build learning — generating ideas, retrieving knowledge, structuring arguments and making judgements. If these are removed, then learning is removed, and human development is diminished.

This distinction made in Lodge & Loble (2026) is important. AI can be used in ways that support learning or in ways that replace it:

• beneficial use supports lower-order processes such as editing and clarification
• detrimental use replaces the intrinsic work of learning — analysing, synthesising and reasoning

The difference is not technological. It is pedagogical. Beneficial use develops thinking. Detrimental use bypasses it. Here are the central risks explained by the article in summary:

• Students are outsourcing the thinking required to actually learn.
• AI improves task performance but undermines long-term learning.
• AI creates the illusion of understanding without real knowledge.
• Overuse of AI reduces critical thinking and independent judgement.
• AI risks widening the gap between high-performing and disadvantaged students.

Guidance from Australian Catholic University (ACU, 2026) is instructive. Although written for universities, it reinforces that AI should support practice, feedback and revision, but be limited where learners must show their own knowledge, reasoning and judgement. In practice, the implications are straightforward:•

• We will use AI where it improves efficiency and supports learning
• We will not use AI where it replaces the thinking, we expect students to develop
• We will continue to prioritise knowledge, writing and analysis as the core of learning
• We will develop authentication methods to substantiate student work rather than an enforcement model to find AI in students’ work (my addition)

Guidance from Australian Catholic University (ACU, 2026) is instructive. Although written for universities, it reinforces that AI should support practice, feedback and revision, but be limited where learners must show their own knowledge, reasoning and judgement. In practice, the implications are straightforward:

• We will use AI where it improves efficiency and supports learning
• We will not use AI where it replaces the thinking, we expect students to develop
• We will continue to prioritise knowledge, writing and analysis as the core of learning
• We will develop authentication methods to substantiate student work rather than an enforcement model to find AI in students’ work (my addition)

Writing across the College and AI

Writing is a central focus for our students, and it is critical that if we are to improve student writing, and by extension their ability to think, we must implement writing practices consistent with the principles referred to above. If student learning improves, it is because students are engaging in the cognitive processes of learning and not contracting these out to an AI. Writing is now a whole-school focus rather than something confined to English. Students are writing more frequently, receiving clearer feedback, and are expected to apply the same approaches across subjects. While there is AI embedded in one of our key writing tools – Writer’s Toolbox – it is structured in such a way as to compel students to do the work of thinking and not avoid it.

Key writing strategies at the College include:

• developing formative writing tasks linked to assessment
• delivering targeted writing lessons aligned to summative tasks
• embedding common and purposeful sentence styles and paragraph structures across subjects
• increasing the use of feedback to prompt revision and improvement
• embedding writing more consistently across all curriculum areas
• teaching writing explicitly to students
• providing coaching and professional learning for staff

The work around our students’ writing aligns directly with our Cairns Catholic Education position on AI. If students are going to learn to think, they must practise thinking and not outsource it. Given that writing is one of the primary ways that students develop their thinking; to allow AI to undermine it would be extremely detrimental to their social, moral and intellectual development. Thinking cannot be outsourced. Writing at the College is improving because work expectations are clear, teaching is more explicit, and students are required to practise and refine their work consistently, and to think about the ways they are constructing meaning in their writing. I commend Emma Ledlin, Angelique Breakspear and Ambre Paolacci, who together with our teachers, have been instrumental in developing these initiatives with our students. I leave you with one last quote from Anna Goldsworthy which strikes a somewhat chilling note:

“The widespread adoption of LLMs threatens linguistic biodiversity further. It is not only a question of lexicon but of punctuation, sentence construction, rhetorical stance. We see it in the overuse of dashes; in the “Rule of Three”; in the symmetrical sentence constructions. In the gussied-up way the text sits on the page, like a corpse ready for the open casket” (Goldsworthy, 2026, The Wordsection).

She writes in metaphor describing the effect AI is having on the richness of our broader human dialogue. It is not simply a change in style, but a flattening of it. When language becomes predictable, patterned and uniform, we lose something essential — the individuality of voice, the nuance of thought, and the connection between language and lived experience. If students come to rely on AI generated expression, they risk losing their own capacity to form language as an expression of thinking. And if that happens, we are not just diminishing writing — we are diminishing the person (my note: see how the writing of the AI basically aligns precisely with the author’s point above). 

I urge you to read the articles in the reference list below as we navigate the use of AI in our school.

Co-Pilot Trial

Because of the potential impact of AI on our students, St Stephen’s Catholic College is the second school in the Diocese to undertake the Microsoft Co Pilot 13+ trial. This trial has begun with both staff and students undertaking AI Literacy courses in readiness for a launch for Yr 9 –12 students in term 4. The purpose of this trial (and then full implementation) is to allow students to enhance their learning through the responsible and ethical use of AI. In term 4 there will be lighthouse teachers using this platform in teaching, learning and assessing. Students will see a modification to task sheets to identify the level of AI use in the assessment for that term – at this stage all assessments are “No AI Use”. We hope this trial will allow our students to improve their correct use of AI and improve their learning outcomes.

2027 Timetable – Proposed Changes

We are proposing some changes to our timetable structure in 2027 with the following intent:

• make the day more consistent and manageable
• provide accessible blocks of learning time
• strengthen pastoral care
• maintain current start/finish times (buses unchanged)

• add flexibility into the structure to be able to support teacher placement on classes and greater allocation of subjects across any two-week period

   

Proposed structure:

• 15-minute Home Form each morning
• five 60-minute lessons per day
• two 30-minute breaks per day
• 10-day timetable cycle

Further updates will follow as this is refined.

ACER Review - Dates: 17–21 August

The review will:

• provide independent, evidence-based feedback from ACER
• examine teaching, learning, leadership and outcomes
• include the perspectives of students, staff and families

Parent voice is part of this process. Further information will be provided.

***Open Afternoon – Time Change***

Monday 1st September 2026 

3:15 pm – 5:15 pm

Reference List

Lodge, J. M., & Loble, L. (2026). Artificial intelligence, cognitive offloading and implications for education. University of Technology Sydney. https://doi.org/10.71741/4pyxmbnjaq.31302475

Catholic Education Diocese of Cairns. (2025). Generative AI Position Statement.

Pope Leo XIV. (2026). Magnifica humanitas: On safeguarding the human person in the time of artificialintelligence. Vatican City. https://www.vatican.va/content/leoxiv/en/encyclicals/documents/20260515-magnifica-humanitas.html

Goldsworthy, A. (2026). The God we made: The threat and promise of artificial intelligence (QuarterlyEssay 102). Quarterly Essay. https://www.quarterlyessay.com.au/essay/2026/06/the-god-we-made

Australian Catholic University. (2026). Artificial intelligence in education guidance.https://staff.acu.edu.au/our_university/research/research-services/generative-ai