The VET sector has a digital capability problem. Not because people aren't paying attention, if anything, there's more conversation about digital skills now than at any point in the last decade. The problem is that much of that conversation is still being shaped by frameworks that describe a digital world we no longer live in.

A lot of organisations are still leaning on the Digital Literacy Skills Framework (DLSF), the Australian Digital Capability Framework (ADCF), or earlier versions of DigComp to define what digital capability looks like. And I understand why. They're familiar. They've given us a shared language. They've helped structure thinking around a topic that can otherwise feel vague and unwieldy.

But here's the thing. The Digital Literacy Skills Framework (DLSF) was released as a draft in 2020 and remains in a draft or interim state. While it has been used in limited program contexts, it has not been widely adopted, standardised, or validated at a national scale, and subsequent reviews have recommended reform or replacement. The ADCF was an attempt to Australianise DigComp 2.1, but by 2025 the Australian Government had moved on, recognising DigComp 2.2 as the preferred framework for guiding digital skills and training. However, these frameworks have not been evolving alongside the sector. They've been standing still while the world around them has changed dramatically.

And when your definition of digital capability is out of date, everything built on top of it, your diagnostics, your support strategies, your course design, your claims about learner readiness, is built on shaky ground.

The Old Frameworks Aren't Wrong. They're Incomplete.

To be fair, these tools weren't bad when they were created. They were built for a world where digital capability largely meant using tools correctly, communicating online, managing files, and doing some basic problem solving. That was a reasonable description of the landscape at the time.

It's not a reasonable description anymore. Today, people search with AI, write with AI, assess the reliability of AI outputs (or don't, which is the real problem), work across multiple platforms simultaneously, interact with automated systems they don't fully control, and make constant judgement calls about privacy, risk, and trust. That's not "digital literacy" in the way we used to talk about it. It's something more complex, more contextual, and more demanding. If your framework doesn't capture that shift, you're measuring a version of capability that no longer matches the job.

What DigComp 3.0 Gets Right

DigComp 3.0 isn't a cosmetic refresh. It's a substantive rewrite of what we mean when we say someone is digitally capable. Importantly, it keeps the same structural backbone — five competence areas, 21 competences, eight proficiency levels, so you're not throwing everything out and starting from scratch. But the substance underneath has shifted in ways that actually matter.

The most significant change is that AI is no longer treated as an emerging topic or an optional add-on. It's embedded across the entire framework. That means the expectation isn't just that someone can use a digital tool, but that they can decide when AI is appropriate, evaluate what it produces, understand its limitations and biases, and manage the risks that come with it. Whether we like it or not, that's the reality of how work gets done now. A framework that ignores it isn't measuring digital capability — it's measuring nostalgia.

The second shift is more subtle but equally important. DigComp 3.0 moves the emphasis from task execution to judgement. Older models tend to ask "can this person use the system?" DigComp 3.0 asks a more useful question: "can this person make good decisions in a digital environment?" That shows up in changes like the move away from old-fashioned "netiquette" toward broader responsible communication behaviour, the expansion of programming into computational thinking and automation, and a much stronger focus on identifying problems and capability gaps rather than just completing tasks. It's the difference between clicking the right buttons and knowing why you're clicking them.

The third shift is about risk. DigComp 3.0 treats wellbeing, misinformation, cybersecurity, sustainability, and digital rights not as peripheral concerns but as core components of capability. Because modern digital environments aren't neutral. They're messy, noisy, and sometimes genuinely unsafe. If someone can operate a system but can't recognise when something is off — whether that's a dodgy source, a biased AI output, or a privacy breach — I'd argue they're not actually competent. Not in any way that matters in the real world.

Why This Matters Inside RTOs and TAFEs

It's tempting to treat framework discussions as theoretical. They're not. They shape very practical decisions, and the consequences of getting them wrong are real.

If you're an RTO owner or executive making decisions about capability uplift, your framework defines what you think the problem is. Get the definition wrong and you invest in the wrong things. I've seen organisations spend serious money improving "digital skills" that don't shift outcomes in the workplace, because they were solving a problem that belonged to 2018.

If you're a trainer or assessor, an outdated framework means unclear expectations. And when expectations are unclear, practitioners fill the gap themselves, which means inconsistent delivery, ad hoc approaches to digital tools, and wildly different interpretations of what "good" looks like. Not because anyone is doing the wrong thing, but because no one has clearly defined the right thing using a reference point that reflects current practice.

And if you're in a compliance or quality role, there's a quieter but increasingly important implication. The more our understanding of work evolves, the more pressure there is to demonstrate that training outcomes align with real-world requirements. If the framework underpinning that alignment hasn't kept pace, the question stops being "did the learner complete the training?" and starts being "are they actually equipped for the way this job works now?" Those two things are drifting further apart than most people realise.

Getting the Definition Right Changes What's Possible

This is why the choice of framework matters. Not because DigComp 3.0 is perfect, no framework is, but because it's a more accurate description of how work, learning, and digital environments actually operate today. When the definition is closer to reality, everything downstream gets easier: describing the digital demands of a role, identifying where someone is underprepared, designing learning that translates into actual performance, and supporting people in ways that are targeted rather than generic.

It moves the conversation from "does this person have digital skills?" to "are they equipped for the way this job actually works?" That's a more useful question, and honestly, it's the question we should have been asking all along.

A Note on Perspective

We work in this space. We've built tools that assess and analyse digital capability, and that gives me a perspective that's informed by what we see every day, where organisations struggle, where learners get stuck, and where the gap between framework and reality causes real problems. You could call that a bias. I'd call it lived experience. The pattern is consistent: the closer the framework is to the real world, the easier everything else becomes.

Where This Leaves Us

The sector doesn't lack commitment to digital capability. What it needs is a definition that actually reflects the world our learners are stepping into, one shaped by AI, automation, and a level of digital complexity that older frameworks simply weren't designed for. DigComp 3.0 provides that. Not perfectly, but meaningfully.

So, if improving digital capability is still a priority, and it should be, the question worth sitting with is this: when was the last time you checked whether the framework you're using still describes the way work actually happens?

About the Author

Matt Peachey