AI

Not to state the obvious, but you’re probably already noticing many little (and big!) ways that Artificial Intelligence is changing how we work.

Some of us are fans, and some of us are wary. But wherever you fall, the good news for leaders is that when it comes to leading in an AI world, you don’t need to be a tech expert or even own the AI strategy to lead effectively.

That said, you do need to understand what’s shifting, where the risks lie, and how to guide your team with confidence.

That’s why I sat down with someone who knows a thing or two about AI, and that’s Tim Pearce from Co-Intell AI, who’s had a world of experience not only in integrating AI into business operations so that humans and machines can collaborate effectively, but also as a leader in the corporate world in the UK and Colombia.

I asked him six questions that leaders are asking me a lot. He had some insightful gems of advice for leaders when it comes to navigating the world of AI.

Suzi: AI feels a bit overwhelming. What do I really need to understand if I’m a leader?

Tim: Start by changing your mental model: AI isn’t just a tech project. It’s a people project. You don’t need to master the mechanics of how AI works. But you do need to understand how it changes workflows, how it feeds into decision-making and the way teams can use it to create value.

Three key insights:

• AI accelerates thinking, but your judgement, taste, and experience are still essential, so you’re going to want to slow down enough to exercise and consider those.
• The learning curve need not be steep; just 10 hours of hands-on play can take you from confused to confident. Bring your learning mindset, the one that allows you to be brave enough to not be good yet. Remember, you’re trying something new. Experiment with writing prompts, summarising notes, or exploring tools like ChatGPT.
• Watch how AI is used in other industries, because shifts in other spaces may well ripple into yours, or could highlight possibilities for innovation.

Emerging risks to watch:

• Shadow AI: Consists of unapproved tools used without oversight. Unapproved AI tools used on the side may seem like a good idea but can pose real security and compliance risks. Steer clear and ensure your tech team are across what you’re using.
• Prompt injection: Malicious prompts can hijack or manipulate AI behaviour. Think of it like a phishing email for AI; a prompt hidden inside an email or document can trick an AI tool into following harmful instructions without you realising. Solid policies are essential to make sure AI is used responsibly – like any tool in your business should be
• Data privacy: Never enter confidential or sensitive material into any tool unless it’s approved and secure.

Suzi: Ok, so if I’m a leader, where do I start? What’s a smart way to start using AI in my role?

Tim: Many leaders can and should begin with low-stakes tasks. 

Things like:

• Drafting emails and reports
• Summarising meetings
• Brainstorming ideas

The real leap comes when you bring AI into your own context. Imagine a Kiwi business leader creating an AI workspace with project data, supplier feedback, and regulations. Within minutes, it produces a risk summary, draft proposal, and communication plan aligned with company values. Think of AI helping you to transform context into action where machines hold the data, and humans provide the nuanced thinking that drives the high value results you’re after.

Here are some Do’s and Don’ts:

• Do use approved tools with data control and privacy features.
• Don’t allow sensitive information into open systems.
• Do keep your data organised, structured, clearly labelled and consistent, so AI can understand it, use it correctly and produce the results you want.

Suzi: What are the biggest mistakes you see leaders make when it comes to using AI early on?

Tim: Good question. Here are some of the most common mistakes I see leaders making:

Mistake #1: Treating AI like Google

AI doesn’t work well with vague prompts. It’s not a search engine, it’s more like a smart junior colleague. You need to explain the task, define the role, and show what good looks like.

Try saying:
“You’re my marketing assistant. Write a first draft of XXX promotion (max 300 words) in our brand voice for audience YYY. Use the attached guide as a reference for tone and style. It shows what ‘good’ looks like.”

Mistake #2: Buying before building

Don’t roll out expensive AI tools before your team is ready. Without support, fear and resistance will undoubtably take over.

Instead:

• Run short learning sprints or pilot programmes.
• Share peer examples.
• Be transparent about your own learning curve.

Quick risk checks:

• Have a policy: what’s OK and what’s not?
• Never deploy AI-generated code or links without reviewing.
• Keep a human in the loop for decisions that matter.

Ideally, a qualified human must:

1. Read and understand the AI-generated code or link before using it.
2. Test it in a safe, controlled environment. 
3. Verify that it does only what’s intended and nothing more. 
4. Check that any external link or reference source is legitimate, secure (HTTPS), and appropriate for business use.

Suzi: One thing I’m sitting with is how does leadership need to evolve in the age of AI?

Tim: Yeah, it’s a big one for leaders to consider and is probably on the minds of lots of your clients. The fundamentals remain: deliver outcomes, apply judgment, stay close to the work. What’s new is the need to move faster, adapt quicker, and model how to learn.

Strong AI-era leaders do the following:

• Start small and win early one improved process at a time.
• Ask for feedback daily, not quarterly.
• Focus on 1% improvements. Small gains compound fast.

But they also build new instincts, like knowing when to trust AI and knowing when to override it. And (easier said than done) knowing when to stay grounded in real value, not the hype.

Governance tip? Build risk habits into the workflow:

• Check your sources. Make sure AI information comes from reliable sources before you act on it. 
• Flag sensitive material. Label data as public, internal, or confidential so you know what’s safe to use in AI tools.
• Always keep human review in the loop. As I said earlier, a qualified human should read and confirm AI outputs before they are shared or published.

Suzi: But what if I don’t own the AI strategy? I might be a middle manager or a senior executive who is key, but who isn’t setting the direction with this stuff. Can I still lead well?

Tim: Absolutely. Strategy may sit elsewhere, but you shape how AI is used day-to-day. Your impact is in how work gets done, not what tech gets chosen.

What you can do right now:

• Be clear with your team on what AI may change (and what it won’t).
• Designate a sandpit for people to play with it. Carve out space for low-risk experiments.
• Celebrate improvements in quality of outputs, not just volume.

Smart questions to ask:

• “What tasks drain our time but add little value?”
• “Where could AI help us raise our game?”
• “What outputs need human review before they go out?”

Pro tip: Set up weekly ‘AI wins & risks’ catchups to normalise learning and flag concerns early.

Suzi: How should Kiwi leaders think differently about AI?

Tim: Since coming from the UK, I’ve noticed New Zealand’s culture to be collaborative, pragmatic and resourceful. I think this is a hidden strength in the AI age.

Why it works:

• Collaboration fuels smart partnerships with universities, suppliers, even competitors.
• Pragmatism means faster testing and safer learning.
• Constraints sharpen focus: AI helps small teams punch above their weight.

When it comes to AI tools, size isn’t everything. A Large Language Model (LLM) is like a public library; it has everything, but it’s overwhelming. A Small Language Model (SLM) is like a mobile library van; compact and practical yet filled with what you actually need.

For Kiwi businesses, that analogy is good news. You don’t need a giant system to see results. A smaller, more focused AI model, tuned to your context, can deliver faster wins and greater trust. What to shift:

• From “prove it first” to “test, learn, improve.”
• From “small market” to “global sandbox.”
• From “safe and slow” to “safe and smart.”

You don’t need to be first; you just need to be ready.

My final thoughts having chatted with Tim? If you’re waiting for AI to settle down, you’ll wait forever. The leaders who thrive won’t be the most technical. They’ll be the most curious, adaptable, and human – isn’t that always the case?!

So, roll up your sleeves. Ask better questions. Try the tools. And most of all, stay in the game.