1
RACA Journal: ISSN 1812-772X
+ 27( 0) 82 560 8718
RACA Journal Publication www. refrigerationandaircon. co. za eamonn @ interactmedia. co. za editor ' s note
AUTOMATION WITHOUT SOUL
Why overreliance on AI threatens trust and human connection.
attended a conference on ventilation technology recently, and
I while it was emphatically not about artificial intelligence( AI) it was notable that AI cropped up everywhere. Coincidentally, before the conference I chatted to an engineer who brought up some of the following ideas, and the opening presentation was on a similar theme.
AI has arrived with the force of every great technological shift before it – only faster, more pervasive and far more intimate. Tools like ChatGPT and Microsoft Copilot are no longer novelties; they are becoming embedded in how we write, design, analyse and even communicate. The promise is extraordinary. The risk, however, is quietly growing alongside it.
At the heart of the current AI wave is a simple but often overlooked truth: these systems are tools – powerful ones, but tools nonetheless. As the keynote speaker argued in her work on human – machine collaboration, businesses’ real competitive advantage does not come from technology alone. It comes from the combination of technology and people. Lose sight of that balance, and organisations risk“ automating away the soul” of what makes them effective in the first place.
One of the most immediate dangers is the erosion of humanto-human interaction. As AI becomes more capable of drafting emails, generating reports and even simulating empathy, it becomes tempting to substitute machine interaction for human engagement. This is efficient, but also deeply problematic. Machines can mimic empathy; they cannot feel it. They can generate responses that sound thoughtful, but they lack lived experience, intuition and moral judgment.
The result is what she described as“ artificial intimacy” – a growing reliance on interactions that feel real but are fundamentally hollow. Over time, this risks weakening the very fabric of collaboration. Engineering innovation has always depended on human exchange: disagreement, spontaneity, shared experience. If those are replaced by polished, AI-generated outputs, something essential is lost. There is a second, more subtle danger: the erosion of trust. As my engineer friend pointed out, with AI-generated content becoming more prevalent, a creeping scepticism begins to take hold. We are already seeing it – articles that feel formulaic, images that appear flawless yet strangely artificial, communication that is technically correct but emotionally flat – even podcasts that seem
RACA Journal I July 2026 real except that the speaker never hesitates nor stutters. As this proliferates, people begin to ask a fundamental question: Was this created by a human – or a machine?
And when that question becomes constant, trust begins to fracture.
If everything might be AI-generated, then nothing is fully reliable. The credibility of published material – whether journalism, marketing or professional communication – starts to weaken. In a world saturated with synthetic content, authenticity becomes harder to prove and easier to doubt. Ironically, the exact tools designed to enhance productivity may end up undermining confidence in the outputs they help create.
This is not a distant, theoretical risk. It is already visible in the homogenisation of content across platforms. When everyone uses the same tools, outputs begin to look and sound the same, riddled with words like‘ resilient’,‘ pivot’ and‘ transformative’. Differentiation fades. Breakthrough ideas risk being overlooked because they are presented in the same tone, structure and style as everything else.
The implications for organisations are significant. Competitive advantage will not come from access to AI – because everyone has access. It will come from how it is used, and more importantly, what is not handed over to it.
Human qualities – judgment, ethics, taste, emotional intelligence – become more valuable, not less. These are not soft skills; they are the defining capabilities that machines cannot replicate. An engineer’ s intuition, a leader’ s sense of responsibility, a communicator’ s ability to connect with an audience – these remain fundamentally human domains.
There is also a broader organisational risk. In the rush to adopt AI, many companies focus almost exclusively on tools: which platform to use, how to deploy it, how to scale it. But this is the wrong starting point. AI adoption is not primarily a technological transformation – it is a behavioural one. Without the right mindset, communication and cultural alignment, even the most advanced tools will fail to deliver meaningful value. Perhaps the most important discipline, then, is restraint. Not every task should be automated. Not every interaction should be optimised. Organisations must decide deliberately where AI adds value, and where human involvement is essential. These‘ human moments’ are not inefficiencies; they are where trust is built, relationships are formed and meaning is created. RACA
www. refrigerationandaircon. co. za
1