Interview with Kate Minogue

‘AI has moved us from “why would I buy?” to “why would I build?” to “I’m heavily dependent on third parties. That creates new risks…’

Interview with Kate Minogue

This interview series brings together voices from across our GenAI Masterclass programme and the wider AI ecosystem connected to the RDI Hub. We are speaking with people who have joined us as speakers, contributors, or practitioners, and who are applying AI day to day in real organisations. Their perspectives are shaped by experience, not theory, and by what actually happens once the demos are over and the work begins.

Each interview is designed as a practical, educational piece, focused on real‑world application rather than hype. Our aim is to give SME leaders and corporate decision‑makers clear, experience‑led insight into what genuinely works with AI, where the challenges lie, and how to approach adoption in a way that is grounded, responsible, and useful.

_____________________________________________________________________________

Speaker bio:

Kate builds and scales consumer AI products, turning complex opportunities into commercially strong, trusted outcomes. With a background spanning consumer tech, gaming and AI, she is known for bringing clarity at inflection points and solving problems teams have yet to define. Her expertise sits at the intersection of product strategy, commercial leadership and organisational execution, with deep fluency in AI, data and consumer behaviour. She has led teams across startups and Meta’s multi-billion-dollar consumer apps business, aligning product, engineering, commercial and marketing to drive measurable growth.

Today, Kate works as a Fractional GM and CPO, provides AI strategy advice to senior leaders, and owns a digital course called The AI Leadership Lab. Previously, she held a senior role at Meta, leading marketing science and data product strategy across EMEA during periods of major industry change. She has taken products from concept to beta, built AI and product roadmaps, and developed high-performing teams and operating models. Her leadership is grounded in clarity, integrity and growth, with a focus on building products that meet real human needs, earn trust, and deliver sustainable commercial results.

_____________________________________________________________________________

A company with no AI maturity wants to start on Monday morning. What do they actually do first?

The first thing I always ask companies to do is look at the tech stack they already have.

Don’t go searching for new tools if you don’t need to. For example, if you’re a Microsoft organisation using the full Microsoft 365 suite, start by exploring the AI capabilities within that. It will be much easier to adopt because it fits your existing systems and contracts.

Next, consider where you should be using AI. Ask yourself:

• What are the biggest pain points in the business?

• Or, if you take a more positive angle, if you were hiring someone tomorrow with a completely new skill set, what would you want them to do?

You can find opportunities in both areas: what needs fixing and what’s missing. These starting points help you focus on real use cases rather than novelty, which is often where companies begin.

What separates a use case that demos well from one that survives real operations?

Use cases that demo well often rely on surprise and novelty. People get excited by something new or unexpected, but that doesn’t always translate into real business value or fit with how work actually happens.

Another issue is scale. A demo often doesn’t reveal edge cases or the challenges of running something across an entire organisation. What works in isolation may not hold up once you increase usage, complexity, or demand.

The use cases that succeed in real operations tend to be less flashy. They:

•        Solve real problems

•        Fit into day to day workflows and are designed around workflows rather than individual tasks.

•        Deliver clear benefits

•        Have been thought through in terms of scaling and potential failures

Often, the reason companies struggle to move from demo to reality is because they choose use cases for different reasons, such as what is technically possible.

How should a corporate think about build versus buy versus configure in 2026?

I often frame this as similar to an outsourcing decision from 5–10 years ago. The first question is: are we comfortable outsourcing core capabilities? That leads to considerations like how important customisation is, how much domain expertise is required, and whether this affects your competitive advantage.

The second lens is your internal technical capability. Can you realistically build something better, and fast enough, compared to what already exists? With generative AI, the answer is often no. Unlike traditional machine learning, where building in-house made sense, today’s large language models rely on scale and capital that only a few organisations have.

The third lens, which is particularly relevant in 2026, is risk and control. AI has moved us from “why would I buy?” to “why would I build?” to “I’m heavily dependent on third parties.” That creates new risks:

• A vendor could disappear

• A product could become unavailable due to regulation or geopolitics

• A system could experience outages

• Pricing models could change

So the key question becomes: which risks can your organisation absorb? Optionality and resilience now matter more than ever when deciding between build, buy or configure.

If advising a CFO sponsoring their first AI project, what’s the one thing you tell them?

Understand the true cost of AI. Early on, there was a perception that AI was cheap, especially compared to people. That’s no longer the case.

Running AI, particularly agents, can be expensive. In some scenarios, junior employees are actually cheaper than AI for certain tasks. Costs were also unclear early on, as pricing models were still evolving and often subsidised.

CFOs need to look beyond surface level costs and consider:

• The cost of scaling from pilot to production

• Employee disruption and change management

• Ongoing operational costs

They also need to move beyond cost cutting as the primary success metric. Instead, focus on outcomes. What are you trying to achieve, and why does this investment make sense? The CFO should challenge assumptions and ensure the business case is grounded in reality.

What does “good” look like 12 months into an AI journey?

After 12 months, you should know things you didn’t know at the start. You should be better equipped to make decisions.

Success is not about adoption metrics like the number of tools or users. It’s about learning:

• How AI works within your organisation

• What you would do differently if starting again

• Whether AI has supported your strategic goals and KPIs

If you’ve spent a year running pilots and you’ve learned nothing meaningful, that’s a failure. But if every initiative is designed to generate insight, then you’re in a strong position to make bigger decisions going forward.

What’s one AI tool you can’t be without?

Recently, the tool I’ve been using most is Postiv (postiv.ai), a LinkedIn-focused AI coach.

It analyses your past posts and content, lets you define your tone and goals, and even considers others in your industry. It doesn’t just help with writing, it also:

• Suggests ideas based on current news

• Generates assets like images and carousels

• Offers alternative ways to structure content

I still prefer to write my own posts, but having a tool that prompts different angles and ideas has really helped improve my output over time and it has saved me time in a task that shouldn't take so much.

What EU AI Act or governance traps are companies walking into without realising?

A common mistake is assuming regulation might not apply or will be delayed. Instead, organisations should start with the intent of the regulation.

One practical step is defining “safe zones” and “off-limits zones” for AI within your business. There’s a tendency to try to use AI everywhere, but some areas carry higher risk.

For example:

• Certain HR functions

• Medical or sensitive decision-making areas

Rather than applying AI everywhere and fixing issues later, companies should define risk boundaries upfront and encourage teams to think in terms of risk from the start.

What did you believe about enterprise AI a year ago that you no longer believe?

Coming from a data background, I used to strongly believe you couldn’t have an AI strategy without a solid data strategy, and that you couldn’t skip steps.

For example:

• You needed good data before basic AI

• You needed basic AI before considering agents

I don’t believe that anymore. Recent developments, particularly in agent-based systems, show that some steps can be skipped.

That doesn’t apply everywhere, especially in high-risk areas. Data is still the foundation of AI and so shouldn’t be completely overlooked but in many cases organisations can move directly to more advanced capabilities without completing every stage of digital transformation.

This is encouraging for companies that feel behind. You don’t have to get everything perfect before starting. In many cases, you can still access meaningful value from AI, even if earlier transformation steps are incomplete.

Three key takeaways:

·         Start with reality, not novelty: Build on your existing tech stack and focus on real business pain points or gaps, not shiny but disconnected AI use cases.

·         Scale and cost change everything: What works in a demo often fails in reality, and AI is not as cheap or straightforward as early hype suggested. Decisions on build, buy, or configure now require careful thinking about cost, risk, and dependency.

·         Learning matters more than adoption: After 12 months, success is not how much AI you’ve deployed, but how much you’ve learned about your organisation and where AI genuinely adds value.

Kate Minogue offers a digital course called the The AI Leadership Lab: A Masterclass for Business Decision-Makers - Empowering Senior Business Leaders and Decision Makers to Navigate AI with Confidence. 

You can reach out to her on her Linkedin here for more details.