Synthesised transcript from May 2026 soft launch webinars

by admin
on May 16, 2026

The transcript below has been synthesised from the two soft launch webinars hosted on 13 and 14 May 2026.

Welcome and Overview — Katie Ford (Microsoft)

Katie Ford: Good morning, everyone. Thank you so much for joining us this morning, post-budget, for the soft launch of the Castlereagh Statement. We aim to be a bit more engaging than the federal budget, let’s say, and a little bit more digestible in terms of content, so we hope you enjoy the lineup of wonderful speakers we’ve got today.

I’m Katie Ford from Microsoft, one of the conveners of the Castlereagh Statement. Danny Liu from the University of Sydney and Jason Lodge from UQ are also on the call.

Before we kick off, I just wanted to acknowledge the traditional custodians of the lands from which we’re all dialling in today, and pay my respects to Elders past and present. I am dialling in from the lands of the Gadigal and Bidjigal people of the Eora Nation, here in Sydney. I also want to acknowledge that some of the most beautiful and powerful thinking about AI and the age of education is coming out of our First Nations community. Traditional knowledge systems are very thoughtful about how they put kin and country at the centre of everything they do, and there’s much we can learn from that.

So, what is the Castlereagh Statement? You’ve probably heard a little about it from colleagues, and perhaps on LinkedIn. Before Easter this year, we released, softly and quietly for feedback, the statement, which is titled the Castlereagh Statement: a cross-sector call to action on Australian Education and Training in the Age of AI. It is very intentionally a cross-sector vision — we have tried to bring together some of the amazing thinkers we’re lucky to have here in Australia, from K-12, from vocational education, from higher ed, from industry (including certifying bodies and industry representatives), and from government.

The genesis is that many of us have been having conversations around: what does AI mean for education and training? This is probably the most consequential technology we’ll see in our lives, and it has big repercussions for society, for work, and for being in this age. As a group, we came together to say: the education and training systems we have in Australia have a lot of strengths, but there are parts that are potentially quite maladapted to this future. So we wanted to ask — what are we tilting towards? What is the future state of where we see education and training? What steps do we need to take to get there? And most importantly, how do we work together, collaboratively, in a coordinated way, to have maximum impact?

We’ve launched it as a green paper because we see it as very much a living, breathing document. We’re trying to build on the shoulders of many great reports that have come out previously. So much has been written about AI in education, AI in skills and training — yes, and we absolutely acknowledge a lot of wonderful work has been done and is being done. We are very lucky in Australia that we have some of the best thinkers in the world in this space. We are not putting this out in a vacuum, and we’re not here to create more noise at all.

The Achilles heel that we see, and one of the key problems we’re trying to address through the statement, is the coordination, the joined-up action opportunity. There’s a patchwork of wonderful activity, but the opportunity here is a more joined-up, coordinated approach. The statement builds on the 2019 Alice Springs (Mparntwe) Education Declaration, which is a really powerful statement about how we ensure Australians become confident, successful, lifelong learners — thinking about equity and excellence in the context of education, with equity being a really important motivator.

We won’t accelerate if we’re doing things in isolation in our own organisations at small scale. We need to think much more boldly and ambitiously about coordinated action in this space. We also acknowledge that the technology is moving rapidly — even since October last year, the pace of change has not slowed down. It is very hard for any organisation, but education and training systems and government in particular, to keep up. Our processes for generating and dynamically adapting curriculum operate at a different pace and scale, so we wanted to think very carefully about how we do this differently going forward. We’ve also heard a lot from industry about the skills shift that’s happening — our economy is transforming, and we want to think very carefully about that.

So in October last year, we brought together around 60 to 80 people from over 30 different organisations — from K-12, vocational education, higher ed, industry (including the accrediting bodies, employers, and employee representatives), the technology companies who build some of this, and key people in government as well. A very motley crew, super diverse, but the idea was to bring people together who can actually start to define what is this future we’re leaning towards. What future do we want for Australia and for students in our country?

Thank you to Danny and Jason in particular, who took a lot of wonderful feedback from many on this call and many others, and tried to condense it down into something simpler: three goals, six principles, and three horizons that we’re working towards.

The three goals

We won’t go through this in detail, but in summary:

  1. We need to be clear about what we value in education and in learners, and ensure our metrics — our ways of measuring this — are well aligned. Rather than continuing, in some instances, to measure what’s easiest to measure, we need to align measurement so that we achieve these outcomes and focus on the right things.
  2. We don’t yet have an agreed framework for what future-ready learning looks like from kindergarten right through to adult lifelong education. What is age-appropriate learning along that pathway, and how does it align with changing societal and workplace needs?
  3. Ensuring that every Australian — and this is really important, because equity matters here — is able to confidently, critically, and creatively engage with AI. We don’t want people left behind.

The six principles

Very briefly, the six principles are:

  1. Define the future-ready capabilities. We spent a lot of time talking about not just AI skills, but the human qualities that will matter — learning how to learn, and how to use AI competently, creatively, and critically.
  2. Let go of things that may need to be changed, retired, or adapted. Everyone is learning here — nobody has all the answers — so humility matters.
  3. Reconceptualise learning and assessment. Australia is already a global leader on AI and academic integrity, but the evolution from content to process, to ensure we’re adapted to this age and measuring the right things well, has further to go.
  4. An agile, capability-focused curriculum. A huge piece of work across K-12, VET, and higher education.
  5. Empower educators. Teachers and teacher agency are at the centre of this, and we need to make sure teachers are given the opportunity and support to learn and use this well in their own work.
  6. Pedagogically aligned AI. We care about using technology in a way that enhances and preserves learning, not displacing it.

The three horizons

We’ve started — but not yet finished — thinking about three horizons. In the first horizon, how do we stabilise? Educator capability uplift, responsible use, and assessment change. In the medium horizon, how do we think about the whole curriculum piece, and how do we align incentive structures? And then the further horizon is some of the bigger, bolder things — no less important — where we’d love to work with you on how we co-design that future together.

This is what we all believe in — what we must do as a country. We’ve really worked hard to put this together in a green-paper-like fashion to get your feedback. Do these principles and goals resonate? And what do you think we actually should do together, and in our own individual organisations and spheres of influence, to effect this change?

I’d now love to hand over to six incredible people, each from very different backgrounds, to share their perspective.

K–12 Perspective — Bridget Pearce (Brisbane Grammar School)

Bridget Pearce: Thank you so much for inviting me to speak. I’m really privileged to represent the K–12 sector. I’m a teacher and pedagogical coach at Brisbane Grammar School, although the views I express are my own and are about Australian education as a whole. I also speak as a parent with two children in primary school, and as a Master of Education student at QUT.

Let me start with a story. Picture this. It’s Year 8 camp. Groups must pitch a tent without instructions. The sun beats down, students are sticky with sweat and sunscreen, and they groan at their puzzling piles of poles and pegs. After much frustration, reality sinks in. No one is coming to their rescue. If they want something built, they must figure out how to build it and work together to build it.

I appreciate everyone has AI-in-education-metaphor fatigue. I’m sorry to say, I will return to this story.

Prior to convening on Castlereagh Street, Vince Wall phoned me. He said there might not be many teachers from K–12 in that room in Sydney, so it was incumbent on us to represent more voices than our own. So we went off and started talking to people. My generous mentor, Dr Hannah Campos-Remon, suggested that when I have these conversations, I start with the question: what do you see as the purpose of education?

Almost unanimously, respondents talked about developing people of character who have the capabilities, knowledge, and dispositions to make meaningful contributions to their communities.

If there is so much consensus, why do we need to get together from different states and sectors to calibrate? Because we have an alignment problem. There is misalignment between the capabilities, knowledge, and dispositions that matter now, and those our existing learning and assessment systems incentivise.

This is what I see:

  • Our communities need innovative problem solvers, but our grading systems reward those who follow recipes.
  • Our communities need systems thinkers who connect disparate dots, but our marking guides call for siloed, subject-specific content recall and application.
  • Our communities need collaborators with interpersonal skills, but the crowded curriculum sidelines relational pedagogy to plow through the assessed content.
  • Our communities need people with confidence, compassion, and awareness of who they are and where they’ve come from. That sounds like a job for the humanities subjects, but tertiary admission scaling mechanisms drive students away from choosing them.
  • Our communities need ethical thinkers who can drive, create with, and respond to AI critically and with commitment to the common good. But nationally, our approaches to AI are ad hoc, uneven, and they routinely incentivise concealment and dishonesty over integrity.

You know, most people who read Frankenstein blame the creator for the monster.

We need a national consensus on the knowledge, capabilities, and dispositions that we want humans to possess, and aligned assessments capable of measuring them. Sometimes when I say we need to focus more on dispositions or capabilities, people think I am saying knowledge doesn’t matter. We are not saying that. And that type of binary bias restricts the conversations we need to have to move forward in the best interests of our students. Domain knowledge will continue to matter. But we need to get the balance better to make school more meaningful, motivating, and empowering.

This is not a new idea. Australia has incredible pioneers in this space, but the idea has not thrown down roots in mainstream Australian education. It needs to.

So, what next? The Castlereagh Statement takes a systems perspective, one of the key concepts of which is emergence. The path forward will take shape as we walk it, responsive to the interactions that need to happen now.

People from schools: we need you to table this statement in your communities of practice. When your curriculum authority asks for feedback, please give it with kindness. It doesn’t matter if you were in that room on Castlereagh Street — if you agree with this statement, you are encouraged to build consensus and cooperation within your sphere of influence.

Leaders from government and higher ed: we need you to believe in teachers. We care about nothing more than optimising our students’ impact on their communities, but our capacity to do that is stifled by the systems we’ve inherited. There is a misconception that teachers are resistant to change. It’s not true — we’ve never knocked back a pay rise. We are resistant to bad change. We’ve grown weary of fads and change that comes with more work, but no time to do it. We need you to give us the time, tools, training, and trust to be successful partners in the processes outlined in the Three Horizon Framework for Action.

Those Year 8 boys built their tents. They experimented and shared what worked and didn’t. They persisted with their goal, despite the ambiguity, and they figured out the architecture together. We convened in that room on Castlereagh Street because the reality had sunk in: no one was coming to our rescue. I wonder if you are here today because you’ve realised that too. We need to build a version of education that empowers people to thrive in the time we are in. We must figure out how to build it, and work together to build it.

Thank you.

Industry & Accrediting Body Perspectives

Day 1 featured Catriona Lavermicocca from Chartered Accountants Australia and New Zealand; Day 2 featured Helen Fairweather from Engineers Australia. Both perspectives are reproduced in full below.

Day 1 — Catriona Lavermicocca (Chartered Accountants Australia and New Zealand)

Catriona Lavermicocca: Thank you, Katie and Danny, and everyone for the opportunity to speak.

Artificial intelligence represents one of the most significant shifts the accounting profession has faced. From CA ANZ’s perspective, AI brings enormous opportunity — greater productivity, improved quality, and the ability for accountants to focus on high-value advisory work. But it also brings real risks if future professionals are not equipped to use AI ethically, critically, and with professional judgment.

As a professional accounting body, CA ANZ has responsibility across the entire education and capability pipeline — from advocating for strong foundations across K–12, vocational, and higher education, through to preparing future chartered accountants in our own graduate diploma and CA program, and ongoing continuing professional development. This aligns closely with the growing national view, reflected in initiatives like the Castlereagh Statement, that preparing graduates for an AI-enabled workforce requires connected, system-wide thinking — not isolated fixes.

Opportunity and risk: a shared national challenge

AI is already reshaping how value is created in accounting. Routine tasks can be automated, analysis accelerated, and decision-making supported at scale. But at the same time, we have seen highly visible failures — fabricated references, unchecked AI outputs, and errors in professional contexts — that remind us AI can undermine public trust when it’s not properly governed.

This tension is not unique to accounting; it’s part of the broader challenge recognised nationally — that AI amplifies both opportunity and risk faster than our education systems are designed to respond. The Castlereagh Statement frames this clearly: it’s not AI itself that creates the risk, but fragmented responses and unclear expectations of human responsibility.

For CA ANZ, the principle is simple: AI does not replace ethics, accountability, or judgment. Being work-ready in an AI-enabled profession is not about knowing how to use the tool — it’s about knowing when to question it, when not to rely on it, and when to override it.

What “AI work-ready” means

Across our engagement with educators and employers, we see work readiness resting on three integrated capabilities:

  1. Technology and AI fluency — understanding what AI can do, where it adds value, and where it introduces risk.
  2. Human capability — critical thinking, professional skepticism, and judgment. These remain central to professional practice, and are increasingly important when working with powerful tools.
  3. Ethics and integrity — the ability to act transparently, responsibly, and in the public interest.

This mirrors the thinking behind the Castlereagh Statement’s call for a shared national view of what we value in human learners and professionals. These capabilities cannot be bolted on at the end of a degree or designation. They must be built progressively across learning pathways, from early education through professional preparation. Without ethics, fluency creates risk. Without judgment, automation creates false confidence.

Challenges for education and assessment

This creates deep challenges. Evidence shows learners can now produce technically polished work using AI without developing deep understanding — what is often described as cognitive underloading — and this raises difficult questions across all education sectors. If traditional assessment primarily tests recall or procedural accuracy, and cannot evidence ethical judgment or human oversight, then we risk certifying graduates who look capable on paper, but are not ready for professional responsibility.

This challenge is explicitly recognised in the Castlereagh Statement, which calls for a shift away from detection-based integrity models, and towards assessment that values learning process, judgment, and capability over output alone. For CA ANZ, these are not abstract education debates — they go directly to competence, public trust, and professional accountability.

We advocate for coherent education pathways across schools, VET, and higher education, consistent with the Statement’s call for connected, scaffolded learning. At the same time, we take direct responsibility for the design of our own programs. Future accountants will use AI at work, but they must be assessed on their ability to use it responsibly.

In closing: AI makes the profession more capable, but it also raises the stakes. By aligning professional education with the broader national vision articulated in the Castlereagh Statement, CA ANZ is focused on ensuring future accountants are not just job-ready, but ready for the work of the future — work that demands judgment, ethics, and trust, as much as technical skill. Thank you.

Day 2 — Helen Fairweather (Engineers Australia)

Helen Fairweather: Thank you, Katie. I’d reiterate, to start with, the importance of having every sector in the room, and I was delighted to be invited along to the summit. It’s not very often that you get to participate in something that became clear to me is going to be a really important process — to start unifying what we’re doing in Australia and laying the foundations for a future that is going to be so important for all of our students. And I think the thing that unified us all was the passion that everybody in that room had for achieving that.

I’d like to point out that our professional accreditation in engineering is outcomes-based. So we’re not regulators, and we’re not prescriptive. We don’t specify what educators have to teach, or which tools they need to use. Our central role is to provide a level of assurance that graduates have the capabilities expected of a professional engineer — regardless of how learning occurs, or how the technology evolves.

I hope I’m not going to embarrass Danny, but I do want to relay how I got to here. I was really struggling with: what am I going to do in accreditation about generative AI, and how do we deal with that to assure those graduate outcomes? Danny very generously flew up to Brisbane from Sydney and gave up 3 hours — probably 8 hours of his time by the time you factor in all the travel — to not present to my accreditation board, but to facilitate a discussion with the members of that board. It was a really pivotal moment, I think, for the board members, and the way Danny approached it was central to that pivotal turning point. He didn’t tell us anything, really. He just asked a series of really insightful, carefully framed questions.

That then led to my participation in the National Education Summit, and it was genuinely one of the most rewarding two days I’ve had in this space. Probably because I was coming from this position of what am I going to do? — and then to get in a room with all of those people who are so passionate about doing something. There were some tough discussions, and there was no silver bullet, and no illusion that a blueprint would emerge. But it appeared to work. One of the things we were asked to do was to resist pretending we already know where we are going, because we clearly don’t.

There was some consistent framing across the very different sectors, but there is quite a different approach and point in the journey that the different sectors are at — Claire from the VET sector will talk about that, which was really an eye-opener to me to understand where the different sectors are at.

We started with three provocations at the start of the summit to help prime our thinking, and they were intentionally uncomfortable — asking whether we are here to save education as we know it, or imagine what comes after it. Are we redesigning the system from scratch?

One of the analogies that was given was the AI gym analogy, and that resonated strongly with me. You don’t build strength by watching somebody else lift the weights. If we allow our thinking to atrophy, then outcomes will suffer — regardless of how polished those outputs appear. As a more mature woman, that really resonated with me, because of the worry that I have about continually cognitively offloading a whole lot of thinking that I would have been doing previously to generative AI, and what that is doing to my ability to continue to think as I’m moving to an older age bracket.

But not only from that personal point of view — from an outcomes-based accreditation perspective, that really matters deeply. Capability is everything, because the capability of engineering graduates is not just about producing answers; it’s about judgment, responsibility, and knowing when not to rely on the tool.

I’ll finish with this: accreditation doesn’t stand above the sector — it sits within it. And the Castlereagh Statement matters because it recognises that this moment is about shared stewardship. My experience over the past months engaging with the Castlereagh Statement gives me cautious confidence that, while there are no easy answers, there’s a growing willingness to do the work together. Thanks, Katie.

Vocational Education Perspective — Claire Field

Claire Field: Thank you, Katie, and everyone who is here. I wondered a little bit at first at the ordering — normally it’s schools, VET, higher ed, and then industry at the end. And yet, of course, because he is our resident genius, Danny has got this perfectly right. The VET sector sits perfectly between the work of the school sector (and is involved in schools through VET in schools) and is working closely with the accrediting bodies and with industry.

Coming after two fabulous speakers really signifies some of the challenges that the VET sector has. We’ve got those students coming through from the school sector, we’ve got similar challenges to engineers and other accrediting bodies — how do we certify that the graduates exiting the VET sector have those skills? And obviously, industry is still working this out. It’s quite a movable, shifting challenge for business, and therefore for the tertiary sector.

Sadly, the VET sector is a long way behind both the schools and the higher education sectors. TEQSA was world-leading when it landed its first webinar on generative AI back in February 2023 — because of all the good people in the higher education sector. The schools sector, bringing together all of the states and territories, landed on a framework by October that same year, and have kept that updated a number of times since.

A lot less has been happening in the VET sector at both the national and the systems level. I know there are good providers who have been grappling with this, who’ve been using TEQSA’s resources. The good news is that ASQA is about to release principles on the responsible use of AI that will take us a step into that important discussion about academic integrity. But sadly, I don’t think that is nearly enough. The conversations I’ve had in the sector indicate there are a whole lot of people ready for more — and that is where the Castlereagh Statement will hopefully, with your input, help to guide and shape what response the VET sector brings.

A diverse sector at risk of falling further behind

We’ve got the challenge of keeping pace with how industry is using AI. We have a really diverse group of learners in the VET sector and across communities. We have VET providers working in capital cities with large businesses who are using the cutting edge of AI tools — and we have others educating students in regional or very remote communities from low-SES backgrounds. We risk widening the digital divide if we don’t come together within VET, and across the whole education sector, to really lift what we are doing and keep current. Ultimately, not only do individuals suffer, but productivity across the whole economy suffers too.

We have large numbers of students coming from disadvantaged backgrounds, living in communities where AI use is not as prevalent, in regional areas where employers are not using AI tools. There are real challenges for VET as it thinks about preparing its learners — learners who have been using AI in their schools, properly and with guidance, coming into our sector while we are still right back at that starting point.

“ChatGPT doesn’t hold a hammer” — and why that’s dangerous

I think one reason we’ve been slow in VET is the assumption that ChatGPT can’t hold a hammer, can it? Well, no, it can’t. A whole lot of VET assessment is competency-based — students demonstrating what they have learned. But in having that view, we have potentially created really important systemic risks for people who are working in those very manual occupations.

Take an example. The Certificate III, Certificate IV, and Diploma of Workplace Health and Safety are qualifications that sit in the Business Services Training Package — lots of text-based work to attain those certificates. About 20,000 students per year enrol in those courses. That’s 20,000 in 2023, 24, 25, and another 20,000 this year — we could have 80,000 students out there already. These people go out and have responsibility for health and safety in hospitals, in large aged-care homes, on construction sites. We are yet to have a national conversation on academic integrity, or on rethinking assessment, in those qualifications. So it is serious, even though a lot of VET assessment is competency-based.

Academic integrity isn’t just about cheating on exams. It is actually dangerous in the workplace. We are graduating people and we are not sure what they really know and what they have really learned. I do worry about inequality, and — things that make you not sleep at night — I worry about Australia’s economic productivity. There are ways we can augment the work we do, not replace people; we don’t want the robots taking over, but there are ways we can work differently and lift productivity. We are missing that opportunity at the moment.

Australia risks falling behind

There was a survey released by the Gallup Foundation, done last year, with data published more recently. It looked at the US higher education system, including students from community colleges — what we would call vocational higher education or high-level VET. Asked “as far as you know, what describes your college’s position on student use of AI?”, 10% said they are encouraged to use it as much as possible, and a further 42% said they are encouraged to use it with some limits. That’s 52%. I absolutely guarantee you there is no way that 52% of training providers in Australia have that approach. Others outside Australia are — so we really need to lift our game.

The next wave: wearables

Hat tip to Professor Phil Dawson from the Centre for Research in Assessment and Digital Learning (CRADLE) at Deakin University, who recently posted about wearable AI. They use a beautiful example of people who have vision impairment and learn to use a cane. At first, they are focused on the cane — what it feels like in their hand, how heavy it is, how it gets in the way. Very soon, they use the cane as an extension of themselves — it tells them whether the path ahead is smooth or bumpy, whether there’s anything in the way.

There are millions of people across the world, and here today, who are choosing to wear wearables — not because they’re creepy (maybe some are) but because it helps them think about and engage with the world differently. Do you know if they’re coming to your very practical, very VET-focused, very hands-on competency-based assessment, let alone potentially sitting in your university lecture theatres? I can’t tell you which of those photos in Phil’s post are wearing real glasses and which are wearing AI glasses, because he didn’t share the answer — but it’s a real and present challenge.

There is an enormous number of ideas in the Castlereagh Statement that are focused on VET — ways we can work together as a sector and with the other sectors — but your input is what’s going to be needed to make it work. Thank you.

Higher Education Perspective — Adam Bridgeman (University of Sydney)

Adam Bridgeman: Thanks, Katie, and thanks everybody else. I’m going to repeat a lot of what has been said already, so forgive me. I’m talking from the university perspective — really from my perspective — and I can see there are people from Sydney and a few other universities on the call.

My perspective is that we’re ready to change, but we cannot do this alone. So I’m going to talk about three things: connection, values, and contribution.

Connection

Katie talked about kindergarten to lifelong learning. This is the whole journey for us. What does lifelong learning look like when we might have more spare time if AI is taking jobs? What am I going to be doing in my retirement? What new things is AI going to help me learn?

The biggest risk is fragmentation. Particularly in the university sector, we’re really siloed. We act as self-contained systems. We design curriculum on our own, we assess on our own, we define what success is on our own. We’re fragmented, and we’re not interacting with the school system, the VET system, industry, and our professional bodies as much as we need to. Yes, we’re doing it — and Helen, sometimes we give you 500 pages of accreditation documentation every 7 years and then try to ignore you — but we need to do this a lot more, and a lot more genuinely.

But also, inside universities we are incredibly siloed. People in the same corridor will not be talking about curriculum, assessment, or what student success means. Sometimes when we’re thinking about assessment, it’s hard to get two educators in the same corridor to have a conversation. So we need to work differently. We need to work with the people on this call. We need to work with schools and VET to define those shared capabilities Bridget talked about. We need to partner with industry, treat them as co-educators and experts, and embed real-world learning.

Just hearing what Claire was talking about — we do not need to just go back to exams. That’s not the answer here, not only because of the huge number of those AI glasses being sold, but because we need to have real-world learning and real-world verification of learning. As Claire put it beautifully: AI doesn’t hold a hammer, it doesn’t hold a stethoscope, it doesn’t hold my cat when my cat’s poorly — that’s somebody with a university qualification. We need to know that that person has the skills we say they have, and we can do that most authentically and most securely when they’re actually doing the things we are training them to do.

We need to build pathways where learners can move seamlessly between study and work, back again, and whatever else it will be in 20 years’ time when AI has really impacted the workforce. So we need to think about institutional autonomy, but we also need to think about the system as a whole and the stewardship of it. We can only do that if we all lean in together.

Values and incentives

We must align what we say we value with what we actually reward. AI is forcing us on an uncomfortable journey, answering some uncomfortable but necessary questions. Are we still rewarding what matters?

As Bridget said, we’ve valued in our students the ability to reproduce knowledge — following recipes. The shiny, polished essay or the shiny, polished presentation, over the content. And we’ve valued individual performance. We’ve done that with our students, but we’ve also done it with our workforce — our educators, our academics, our teachers. We’ve valued the things that they do that are shiny, or about research, not only in our recruitment but also in the way we promote.

We’ve got to value the things that are going to be important in the future: judgment, creativity, ethical capability, the ability to adapt and to learn in a lifelong journey. So we have to be ready to redesign assessment, and particularly curriculum, around capability — not just content. Because systems don’t change because ideas change; they change because of the way we incentivise things. That’s why we also need to be talking to government. It’s a collective reset of what we value and what we reward.

That means redesigning assessment around process and growth, and not so much about product. Capability, not just content. And I think relational pedagogy is at the heart of that.

Contribution

This is an open invitation. As Katie said, this is a green paper, it’s not a blueprint. It’s a coalition of the willing to get involved and think about the ideas. For universities, it’s about opening up, sharing learning, talking, and rewarding the things that matter.

I’m going to repeat what Bridget said, because it’s so important: believe in teachers. Believe in teachers and believe in our students. We need to work with industry and treat them as co-educators and experts, and recognise and develop capability in our students. With schools and VET, build that connected, lifelong life cycle. And with government and the regulators, enable the innovation, because policy can drive some of the changes we want.

For us as educators and our students, we need to help shape what comes next, so that it’s being driven by education, by our community — not driven by corporate need. It’s a connected conversation, it’s not parallel reforms. Thinking back to COVID, we worked quite closely together, but then we came back and it’s all gone — we didn’t carry it, we didn’t build on it as we might have done. So I don’t think this is about fixing universities. It’s about building a coherent, lifelong learning system for Australia as a whole. And none of us can do it alone. We all need to do this together.

A real-world authenticity for the new world. Thank you, Katie.

Industry Perspective — Peita Davis (Business Council of Australia)

Peita Davis: Thanks, Katie and everyone. Hi, I’m Peita Davis from the Business Council of Australia. We are the peak body of Australia’s largest businesses, with about 130 members — but in my job I speak to businesses of all sizes. The perspective I bring is, in a sense, that of the end user of your education system. So it’s very much an employer perspective.

What employers expect of graduates today

I’ll start with what employers — large, medium, and small — are telling me are their expectations of the young people they’re hiring today.

The first expectation is that they have incredible human skills: critical thinking, communication, problem solving, teamwork — all those really core skills that perhaps we used to consider were humanities or philosophy. These are exceptional. That’s what they’re asking from graduates.

The second is that these young people have great AI skills — strong AI literacy. We’ve just heard today that the system is not actually supporting the development of those skills, but the expectation of employers is that young people are arriving with them. And I think we need to be really careful not to assume that because they are young, and they’re AI users, that they are AI literate. Literacy means understanding governance, understanding the risks of AI, when and when not to use it — the full spectrum of what AI is and what it can bring.

A lot of small and medium businesses in particular are hiring young people expecting them to teach the rest of the workforce how to use AI. They’re looking to young people to be the answer to this AI conundrum. So there’s a huge amount of expectation on young people. Businesses are not wanting these skills in 5 to 10 years’ time. They are expecting them today.

What’s keeping me up at night

What that means — and it’s keeping me up at night — is that I do not feel our education system, from K–12 all the way through different forms of tertiary, is developing those skills now. Everyone on this call knows that this is a problem, and we’re all trying to solve it. But this is both a top-down and a bottom-up problem.

At the federal level, AI skilling is split across the Department of Education, the Department of Industry, DEWR — none of them is taking full responsibility for AI skilling for Australians. The same plays out at the state level, where states don’t really know how to start embedding AI skills from K–12 all the way through. Everyone is facing financial issues, and AI skilling is just falling further and further down the list of priorities. The fractured nature of this problem — no one really holding it and saying “right, we’re going to solve it” — is the biggest challenge going forward.

From the bottom up, it comes down to capability — the capability of our teachers in every classroom being comfortable enough with AI themselves that they can actually start building it into their curriculum and teaching young people. I remember when we were at the Castlereagh workshop, one of our wonderful participants from the Victorian Department of Education told a story about when they were rolling out electronic whiteboards. Everyone felt they had them; this was going to be the next big productivity uplift for schools. Then they would visit some schools and see those electronic whiteboards tucked away in storerooms, completely unused. Unless we have a bottom-up understanding and groundswell of the importance of AI skills and human skills, kids aren’t going to be taught, businesses will be disappointed, the whole system fails, and Australia will be left behind.

A growing divide

Picking up on the divide that’s happening: my members — big businesses, big corporates — are training their staff in AI. People within those businesses are coming out highly AI-literate, and those skills will be transferable across any job they take. That’s leaving a massive divide with the rest of Australia — both young people, and the current working population, who do not have access to the incredible AI training that some of my members’ staff do.

What we need next

The Castlereagh Statement is a really great first step. But this phenomenal group of people who have come together to all say the same thing — this is important, this is urgent, this must happen — now needs to deliver very clear guidelines and next steps. We need to ensure that every classroom at every stage of schooling has AI literacy embedded — all the way through to VET, to higher ed — and then we also need to provide lifelong learning options for people already in the workforce. We need to come up with really clear steps to deliver to government to say: this is how we solve this problem.

For me, the next step of the Castlereagh process is a really clear list of asks for governments, both federal and state, to say: these are the expectations coming from government, from business. What do we need to do to make sure that our young people are not being left behind? Thanks, Katie.

Future Skills Perspective — Tim Burt (Future Skills Organisation)

Tim Burt: Thanks, Katie. Hi everybody. I’m from the Future Skills Organisation. We’re one of 10 Jobs and Skills Councils funded by the Australian Government.

I just wanted to spend a couple of minutes telling you about a piece of work we’ve completed separately and independently from the Castlereagh Statement, but with really strikingly similar findings. Around Christmas, we were tasked by Mr Giles, the Commonwealth Minister for Skills and Training, to complete a National AI Skills Report, building on the National AI Plan that was released in December last year. Essentially, this is a baseline understanding of the AI skills environment across schools, vocational education and training, higher education, and the non-accredited training space — with a view to identifying what those key priorities for action should be.

We framed the research around four key questions:

  1. What are the skills Australia needs to compete in an AI-driven economy?
  2. Do we have the right skills initiatives to deliver those skills? Where are the gaps?
  3. What is needed to mobilise the nation around AI skills — and what kind of metrics would be useful in measuring AI skills uplift?
  4. How do we ensure First Nations peoples’ considerations meaningfully shape the impact of AI skills initiatives?

We delivered the report on a really short timeline — between January and April, a couple of weeks later than originally anticipated. The four work streams were a literature review and skills/usage scan, a desktop scan of existing AI training initiatives, a targeted consultation with around 300 individuals across 50 different groups (state and territory governments, sector industry peaks, Commonwealth agencies, tech providers, unions, and diversity and inclusion organisations), and a separate First Nations insights piece undertaken by a First Nations-owned research organisation. This is very much a baseline report — it will be a biannual report from here on, with the next due in October/November. Whilst the initial focus was vocational education and training, we’ve been encouraged to think more broadly across schools, higher education, and non-accredited training.

Three key intersection points with the Castlereagh Statement

These projects were undertaken independently, but the findings overlap strikingly. There are three key intersection points:

  1. AI is a workforce and skills challenge — it’s not an IT rollout. It’s a question about people, capability, work, and skills. Human capabilities, domain knowledge, subject matter expertise, the need for human judgment, ethics, human oversight — these continue to matter, fundamentally, and perhaps more than ever.
  2. Coherent pathways from early childhood through lifelong learning, recognised across all sectors, are fundamental. The whole of the education and training system and industry need to come together to smooth that journey. We’re all lifelong learners now. Addressing the fragmentation that Adam succinctly called out is going to be critical.
  3. Coordination is a real constraint. There is a lot of activity happening across Australia in education, training, and AI. How do we align it? How do we scale it? How do we bring all of that work to benefit the nation? The whole intent and purpose of the Castlereagh Statement is critical in facilitating the shared commitment for urgent action.

So now we’re thinking about action. The piece of work was conducted independently of the Castlereagh Statement, but we’ve been having conversations with Danny and the team around how we collaborate much more closely, and use the expertise of this group and others to socialise and develop those priorities for action that Peita called out. We’re really keen to set forward an agenda for action — what the key actions are for government, in partnership with industry, in partnership with the education and training system, in how to uplift AI skills at scale across the economy. Really grateful to be here, really excited for the future opportunity. Thanks, Katie.

Synthesis and Next Steps — Danny Liu (University of Sydney) and Jason Lodge (UQ)

Danny Liu: Thank you so much, everyone, for your perspectives. As Adam and many people were saying, it has to be everything changing at once — people all working together, not in individual silos. As part of Castlereagh, what we’re hoping to achieve through every single person on the call (and listening) is to say: we need to get together and have these collective conversations.

I asked Jason for some quick synthesis thoughts after hearing the six speakers — sorry to put you on the spot, Jason.

Jason Lodge: I think there’s a lot of intersection, obviously, between the different comments. Going back to the opening, that was a big part of the reason we were doing this — because we were all doing it in an isolated way. And it’s good to hear that there seems to be a crystallisation of a common set of ideas that we’re hovering around. That means we’re on track.

Danny Liu: That crystallisation leads to the next steps. As Peita was saying, we now have our general direction. One of the things Jason likes saying is that it’s a compass, not a map. The statement is a compass; what we’re trying to build together now, as a massive group across the country, is the map we can draw out.

When people are signing up for the statement, they’re saying things like: they recognise they’re working in silos, they want to really make a difference, and work together on this. So this is part of the opportunity to do so — a pathway to contribute to this nationwide work.

I’ve popped a Padlet link in the Zoom chat. Over the next little while, please pop on really brief thoughts about what you might do in your own sphere of influence. One of our speakers, Bridget, talked about that sphere of influence that you have. Every single person here can do something to move the conversation, move the action along — we’d love to hear what you might want to do over the next one to two months.

The roadmap: the next five months

What we’re aiming for over the next five-ish months:

  • These two launch webinars (today and yesterday).
  • Sector-specific and cross-sector monthly working parties — these will put together more concrete action plans about the next steps for particular sectors, collect vignettes of practice that we can share across sectors, work on student voice, and refine any parts of the statement.
  • A white paper, hopefully with the support of government, around September/October this year.
  • Potentially leading to a national settlement where we’ll bring everyone together again to have a chat about where we’re at.

We’ll send emails after these webinars — I have your contact details from those who left their information online for us, thank you. We’ll try to get together sector-specific meetups and have some cross-sector conversations too. The idea is action plans, vignettes that are visible, and helping us keep each other accountable. Amanda just posted in the Zoom chat about an ASCILITE business special interest group meeting on Monday — so it’s already happening. It’s great.

What this will look like is still a build-the-plane-as-you-fly-it moment, but we think there should be sector-wide working parties — schools, VET, higher education, and workplace groups — led by people in those sectors but involving everyone on this call who is keen. And cross-cutting those, there will be vertical cross-sector groups: perhaps around IT enablement, around curriculum (aligning it with the needs of employers and community), around credentialing and assessment, and other groups that may form. The point is to get these groups thinking about action plans going forward — Peita’s “next steps” — to then push out to people around us, to government, to anyone who will listen, to say: this is how we need to take the nation forward.

Why this matters

Across the board, we’ve been hearing that our values are misaligned with our measurements. We need much closer connections between everyone working in this space — from the consumers, as Peita was saying, of these graduates, all the way through to the kindy teachers who are preparing their students to be future leaders. If we can have that alignment towards producing humans who are skeptical, trustworthy, creative, adaptable, who have problem-solving abilities and are collaborative, all those things — and also have AI skills — we need to think about whether our current systems are aligned to those eventualities. I don’t think so. So how do we push that forward together?

This is a collective action problem, and we can’t do this unless we all work together — every single person on this call. You are a really key part of what we’re trying to do together, so thank you.

Katie Ford: Thank you, Danny, for some of the most important work here — getting us where we are today, with something coherent that’s able to be put out. It is not a final product. It is the start of collective coordination and collective action. We look forward to engaging with many of you. Please give us very honest feedback. I like the idea of working parties — does that imply we’re going to have a party? We’ll look for opportunities to come together in person and learn from each other.

Categories:

Uncategorized

Tags:

No Tag

Comments are closed