Fill in your details below and one of our AI experts will be in contact to find out how we can support you.
Organisations that invest in AI tools without training their people risk low adoption, misuse and poor return on investment – a problem Datacom's director of AI, Lou Compagnone, calls ‘organ rejection.’
AI literacy in the workplace means knowing what tools are available, understanding your organisation's usage policies and learning foundational skills like prompting – starting with enterprise platforms such as Microsoft Copilot and ChatGPT.
Organisations need formalised AI learning pathways tailored to different roles and skill levels, with AI expected to become a baseline workplace skill comparable to internet or digital literacy.
Organisations across Australasia are pouring money into AI tools, yet many aren't seeing the return they expected. Buying the technology turns out to be the easy part – embedding it into the way people actually work is where most organisations get stuck.
Off the back of her interview on RNZ’s Morning Report, we sat down with Lou Compagnone, Director of AI at Datacom, to talk about what AI literacy really looks like today, why organisations need to get more deliberate about upskilling and what happens when people simply don't want to get on board.
I often call it organ rejection. You can get a new piece of technology, but unless you're thinking about the people and processes around it, you can't expect the whole organisation to use it naturally.
It comes down to people having the means of knowing how to use that tool in their jobs – having the basics, like how to prompt. Otherwise, there's no point in investing in the tool, or people aren't going to be using it safely and it's going to introduce risk to your organisation.
It starts with finding out what tools are available. The obvious ones are platforms like Microsoft Copilot, which integrates into the Microsoft suite – a lot of organisations already have that. And increasingly, we're seeing enterprise accounts for tools like ChatGPT, which is probably the most common one that everyone talks about, and sometimes Claude.
If you're joining a new workplace, it's important to understand what the policies are in terms of what's okay, what's good use of AI and what the business doesn't want you to be doing.
For example, if you're using public AI tools – ChatGPT (OpenAI), Claude (Anthropic), Perplexity – and it's not a company account, you need to be careful about not putting anything sensitive into those models. If it is an enterprise account, you still need to find out what the restrictions are – what your workplace is okay with you putting in. For example, strategic information might be okay, but not customer information.
So, it's about finding out what tools are available and then leveraging any existing training your organisation might have. At Datacom, for example, we have a whole AI literacy training program for our team. We teach the basics of how to use AI, before getting more advanced and providing training in things like how to create agents.
Part of this is a compulsory compliance module that our people must complete before having access to our internal AI platform, Conver. It's a useful starting point, teaching fundamentals like prompting and how to select the right model for the right task – skills that remain relevant and they take forward as their capability grows.
We’ve also recently launched AI in the workplace training, which is free and available for anyone to complete. Hosted via our partnership with learning platform Forage, it’s part of our broader response to the need to build confidence in critical, in-demand tech skills as we foster talent for the future.
My advice: find out what's available for you to use and learn from, and leverage as much as you can.
AI is now building AI. AI companies are using AI agents to build the software that is AI – meaning an agent can take on a whole workflow end to end.
At Datacom, our digital engineering team has developed a series of agents – it’s the start of a significant shift as we move from application driven work to agent-driven operations. Say there's a legacy website that an organisation wants to modernise. A business analyst agent can go into that website, understand how it works, and even draw up paperwork and flows to show how it works.
Across various teams, staff can simply ask for the outcome they need. Agents can automate routine tasks, triage and route work, and complete processes end-to-end. They can read documents, emails, PDFs, and forms and take action. They can also provide real-time insight by surfacing exceptions, risks, and recommended next steps.
The agent becomes the interface. Rather than needing to navigate multiple systems, there’s one consistent way of working, which means reducing manual overhead, increasing productivity, and freeing up time so our people can focus on judgement and value.
I do think there needs to be a more formalised approach. Businesses need to think about what role they want AI to play. Not every person needs the same skillset or tools in their respective role. It's important to consider what you do as a business, what parts of the business need to be upskilled in AI, and then how you're going to roll out the right information to the right people.
We need to get clearer about the role it's going to play, and then have proper learning pathways for different levels, whether you're using, developing or leading AI.
Absolutely. It's changing the way we're going to work forever. The Institute for the Future put out a report in 2025 saying we need to think about AI like electricity or the internet. I heard someone from Microsoft say it's not a technology evolution, it's a revolution. Businesses and people alike should think about AI almost like an industrial revolution where everything is going to change. The whole shape of our organisations, and by extension, our world, will be different forever.
That's why it's so important to be building up the fundamental skills. If you're in a role, think about how you can do more within that role – what can you automate or augment using AI? What does that free you up to do? How does the shape of your role evolve as a result? And where can your skills take you as technology continues to evolve?
It's a great question. I think for organisations that are serious about transforming with AI, it's going to reach a point where AI literacy is simply an essential skill – you're either on the bus or you're not.
