Q&A: From busywork to better work - smarter ways to use AI

• Effective AI use requires a mindset shift from passive question-and-answer to active collaboration, where you direct the tool to challenge your thinking and play a specific role.

• True value comes from adaptation, not mere adoption – redesigning how work is done rather than simply layering AI onto existing processes.

• The biggest barriers to AI uptake are cultural rather than technical, and the remedy is straightforward: start small, align AI to real business outcomes, and build from there.

As part of the Great NZ AI Roadshow, Datacom’s Associate Director – Artificial Intelligence and Automation, Richard Kenyon, is sharing practical insights on how organisations can move beyond basic AI use and start creating real value. We asked for some tips on getting the most out of Copilot, automation and what effective AI adoption really looks like in practice.

A lot of people use AI in the same way they’d use a search engine: they ask a question and accept the answer at face value. That can be useful, but it’s not where the real value sits. 

Before AI, if you had an idea you wanted to test or shape, you would go and talk to someone you trusted – someone who would challenge your assumptions, ask the awkward questions or point out what you’d missed. You can do the same thing with AI. The difference is you have to be explicit about the role you want it to play. 

If you tell the tool, “I want you to act like a sceptical subject‑matter expert” or “challenge this idea and look for the weakest assumption”, you turn it into a thinking partner rather than a question‑and‑answer machine. That mindset shift alone changes how people use these tools.

Prompting comes up a lot in discussions about AI. How should people think about it in practice? 

The first thing is being clear on what you’re actually trying to do. Brainstorming, research and solution design are different activities, and they shouldn’t all happen in one conversation with a gen-AI tool. 
 

If you’re brainstorming, it’s fine to be open‑ended. If you’re researching, you need to be much more disciplined. AI tools are not knowledge bases – they’re trained on past information. If you want current or authoritative information, you should make sure web search is enabled and specify the types of sources you want the tool to use, whether that’s legislation, academic research or trusted industry publications. 

One practical habit that helps is working in stages. Do your brainstorming in one session. Start a new conversation for research. Start another for solution options. At the end, you can ask the AI to summarise those conversations and bring them together. You still own the thinking, but you’ve reduced the effort involved in getting there. 

My advice is to make the work your own, but outsource the labour.

Almost every organisation has at least one person quietly using AI to save themselves time. They’re often very effective, and they’re usually going home a bit less stressed at the end of the day. 

The bigger issue is what happens beyond those individuals. Many people are stuck in the middle – they’re unsure where to start, worried about getting it wrong, or embarrassed to admit they don’t understand the tools. That’s where culture and leadership matter. 

At an organisational level, simply adopting AI tools doesn’t deliver much value on its own. If people automate a task but then spend just as long checking the output, you haven’t really achieved anything. The value comes when organisations adapt the way they work to suit the capability of the tools. 

That means leadership being clear about business goals first, and then asking how AI can help achieve those goals more efficiently or at higher quality. Saying “we need to do AI” isn’t a strategy. Aligning AI to outcomes is.

Adoption is using a new tool. Adaptation is changing how you work because that tool exists. 

A simple example is automation. Many workplaces are full of low‑value manual tasks – copying and pasting information, moving data between documents, sending routine emails. Those tasks don’t differentiate your business, and they’re not particularly satisfying for people to do. 

If AI helps you automate those tasks but everything else stays the same, the returns are limited. If instead you redesign processes so that people spend more time on judgement, problem‑solving and customer interaction, that’s where value is created.

Organisations that get this right lift the work their people are doing. They don’t remove accountability – humans remain responsible for the outcomes – but they remove a lot of unnecessary effort from the process.

Start with the problems you experience every day. 

In small businesses, work often shows up as an overflowing inbox, duplicated effort, or tasks falling between the gaps because everyone is doing multiple roles. Those are good signals that automation could help. 

You don’t need to start with a formal AI strategy. Find someone in the organisation who is curious and willing to try things. Work out what’s costing time or causing friction, and then ask for help – whether that’s from a trusted IT provider, a specialist, or even an AI tool itself. 

One thing I’d stress is being mindful of where your data goes. Make sure you’re using tools that are appropriate for your business and that you’re not unintentionally giving away sensitive information. 

Most importantly, remember that the entry point has never been lower. Much of what took specialist skills a few years ago is now far more accessible.

The honest conversation is that most people can already identify tasks in their day that add very little value – moving documents around, re‑keying information, or managing manual processes. 

When organisations automate that work, they don’t take capability away from people. They create space. That space can be used to grow skills, improve customer outcomes, or focus on work that actually moves the business forward. 

In small businesses especially, roles tend to evolve naturally. People gravitate towards work that suits their strengths once they have time to do it. AI can accelerate that process, but it only works if leaders give people permission to try, learn and adapt. 

This isn’t about removing the human from the loop. It’s about making sure human effort is spent where it adds the most value.

Take one step, even if it feels small. 

It’s never been easier to learn something new, but nothing happens unless you begin. If you’re unsure what AI can help with, you can literally ask it: “What could you help me with in my role?” That’s often a powerful moment for people, because it reframes the tool as a guide rather than something intimidating.

You have three options: do nothing, talk to someone you trust, or ask an AI tool some questions. The last option is often the least judgemental and the easiest to access.

Once people get going, even for five minutes, things start to click. Curiosity goes a long way.

My goal is to show that this is more achievable and more affordable than many people think, and that the biggest shift isn’t technological – it’s how we think about work.

AI is a powerful tool, but it only delivers value when it’s used deliberately, aligned to outcomes, and supported by a willingness to adapt. If people leave with a clearer idea of where AI fits in their organisation, and confidence to take a first step, that’s a good outcome.