Kyle Moffett, Engineering Manager at Rapid7, believes the organisations that succeed with AI will not necessarily be those with the most advanced models. Instead, success will belong to businesses that invest in data quality, human skills, practical literacy and responsible deployment.
Artificial intelligence has rapidly moved from experimental technology to boardroom priority. Across every sector, organisations are evaluating how AI can improve productivity, automate routine work and unlock new opportunities for innovation. Yet despite the enthusiasm, many businesses remain uncertain about where to begin and how to achieve meaningful results.
According to Kyle, Engineering Manager at global cybersecurity leader Rapid7, the conversation around AI often starts in the wrong place.Rather than focusing on models, platforms or the latest technological breakthrough, organisations should start with the problem, not the technology.
"First and foremost, you have to look inward and see what you're trying to solve," he explains. "Are you willing to put forward the investment required, not just financially, but in terms of time, training and organisational focus?"
While the cost of accessing AI technology continues to fall, he argues that the real investment is often cultural. Companies must be prepared to educate employees, create space for experimentation and accept that benefits may take time to materialise.Before deploying AI, organisations should identify which operational processes could potentially be automated, determine their appetite for risk,establish where human oversight must remain part of the workflow and define how success will be measured.
"Fundamentally, how are you going to measure its success?" he asks. "Those answers can only really come from the company itself."
Once those foundations are established, Kyle advocates a practical approach: put the tools into the hands of employees, identify internal champions and invest in training programmes that support adoption.
If there is one area where Kyle believes organisations consistently underestimate the challenge, it is data quality.In his view, data is often more important than the sophistication of the AI model itself.
"The most capable model is often overkill," he says. "I look at it like a Formula One car. Will it go faster? Yes. But do you actually need a Formula One car, and do you have the infrastructure to make use of that capability?"
The comparison highlights a common mistake. Many organisations focus heavily on selecting the most advanced technology while neglecting the quality, governance and structure of the information being fed into it.
Because AI systems are inherently non-deterministic, outcomes can be difficult to predict. Strong implementations therefore require guardrails that guide systems toward desired outcomes and prevent unintended behaviour.
Kyle warns that highly capable models can sometimes become more creative in sourcing information or generating responses. Without controls, that capability may actually move outputs further away from business objectives.For that reason, he argues that data quality should always be the starting point.
The importance of data, he notes, is reflected throughout the industry. AI developers themselves invest heavily in acquiring high-quality datasets because they understand that the quality of inputs directly influences the quality of outputs.
As AI becomes more capable, it may seem logical to assume that technical expertise alone will become increasingly important. Kyle sees a different future emerging.The professionals who will thrive in an AI-enabled workplace, he argues, will combine technical understanding with uniquely human capabilities.
"Creativity and humans' ability to work through ambiguity are key to unlocking AI's power and delivering outcomes," he says.
In software engineering, for example, the strongest performers are no longer simply those with the deepest coding expertise. Increasingly, they are individuals who combine technical excellence with product understanding, customer awareness and strategic thinking.
Designers, product managers and engineers who can collaborate across disciplines are discovering new opportunities to augment their expertise through AI. Rather than replacing human insight, AI is amplifying the value of people who can provide context, judgement and direction. Kyle believes that working effectively with AI will soon become as natural as using a standard operating system.
Understanding how prompts, rules, workflows and AI agents connect together will become an essential workplace competency. However, the real differentiator will remain human judgement.That shift has significant implications for employers, universities and educators.
Kyle believes future talent development should place greater emphasis on outcomes rather than rigid specialisation.
"The people who deliver outcomes aren't necessarily the subject matter experts," he says. "They're often the people who can work across domains and validate outputs."
He argues that organisations should aim to develop individuals with broad capabilities supported by deep specialist expertise. In an AI-enabled economy, possessing a single skill set may no longer be enough.
To illustrate the point, he refers to traditional vocational professions.A mechanic, he notes, is evaluated based on whether the job was completed correctly and efficiently, not whether it was a spanner or a wrench that was used to tighten the bolt. The same principle should increasingly apply in knowledge work.
Rather than focusing solely on how tasks are completed, organisations and educational institutions should be measuring the ability to deliver results, solve problems and apply judgement effectively.Crucially, he believes these skills cannot be learned purely in theory.
"The only way to fully get immersed into that is to immerse yourself into it."
Northern Ireland has frequently discussed its ambitions in AI, but Kyle believes the region is already better positioned than many realise.He points to the success of Northern Ireland's cybersecurity sector, which has spent more than a decade establishing a reputation as a globally recognised centre of excellence.
"We've got the playbook," he says. "It's tried and tested."
The region already benefits from strong university links, established government support structures and a highly skilled technology workforce. Combined with Northern Ireland's strategic position between US and European markets, these strengths create a solid foundation for future growth.However, he believes the ambition should extend beyond becoming an AI hub.
"I think we could dream bigger than just being an AI centre," he says. "We could become an AI-powered workforce."
The objective should be embedding AI literacy across industries ranging from manufacturing and professional services to healthcare and public services. By enabling workers to combine domain expertise with AI-enhanced workflows, productivity gains could be achieved far beyond the technology sector itself.
For Kyle, one of the biggest barriers to progress is a lack of AI literacy.Too many organisations, he argues, acquire AI tools without understanding how to use them effectively.The issue is not access to technology. The issue is understanding how to direct it, manage it and integrate it into meaningful workflows.
While many people are already comfortable using chatbots, the greatest gains come from more advanced, agentic approaches where AI is given clearly defined tasks, guardrails and objectives before operating autonomously within agreed boundaries.
When implemented correctly, this frees professionals to focus on higher-value work requiring creativity, judgement and human interaction.The challenge now is ensuring those capabilities become accessible across the wider economy.
As Northern Ireland continues its journey into the AI era, his message is clear: the biggest opportunity is not the technology itself, but the people who learn how to use it well.
Read the Autumn 2026 edition free online →
Stay connected with NI's tech community:
