Hyphen Perspectives is a series of articles in which our team members share their views on relevant sectors.

Here, Head of Technology DAVID TSCHIRKY looks at the rise of artificial intelligence, how we use it and why moderation is important.

A new industrial revolution is just beginning – and it’s going to be powered by artificial intelligence (AI). That’s not hyperbole because make no mistake about it: this has the potential to change almost every industry on the planet in time.

It’s easy to forget that it is only two years since OpenAI unveiled ChatGPT, which for many people was a watershed moment. To some extent, that marked the opening of Pandora’s box as what was once merely science fiction became science fact as people all over the world learned they can get a detailed result to almost any question within seconds.

Since then, trillions of dollars have flooded into this market as AI innovators jockey for position as the leader. As well as ChatGPT, we now also have Grok, Gemini, Deepseek and a host of other AI platforms based on large-language models (LLMs).

Governments, businesses and other organisations are all grappling with what this brave new world looks like but the rate of evolution makes it difficult for anyone to keep up. This is genuinely disruptive technology and brings so many potential opportunities and threats that even the most informed experts cannot agree on what the specific long-term outcome of our AI infatuation might be for society.

What we should all be able to agree on is that AI in its current form is an efficiency tool for businesses. AI is making jobs easier because it accelerates the work people would typically have to do manually, though it does still rely on verification from humans to get optimal results.

That last point is important because this is, after all, a technology in its infancy and should be treated as such. Take, for example, AI agents. These could potentially replace humans in the workplace as they can be given particularly defined tasks to execute, which can sound exciting or scary, depending on one’s perspective. The reality is that both we as a society and the technology itself are not ready yet, and there is still work that must be done to ensure these agents work harmoniously with humanity.

Platforms can spit out results with factual errors and output often needs to be refined even if the details are correct. There is also the fact the tool is only as useful as the people using it. However, as the technology matures, our collective knowledge and understanding of how best to use it will likely do the same.

At Hyphen, we have sought to capitalise on opportunities and are going through that learning process ourselves. Our team uses AI to varying degrees, with just using it to tailor an email while some in our Design team might use it to create renderings and images.

Speaking from a personal perspective, I can say that I and the rest of the Tech team use it extensively to code, because this is one of the best applications of this technology. Programming language is very specific and there’s so much help available, while it can also autosuggest code. We’re yet to reach that point really when it comes to most generalist, expansive topics, which is perhaps why answers to other questions aren’t always of the highest quality.

We have even set up HyphenGPT, which is hooked up to our data storage and acts as a simple agent. This closed system gathers insights and information that can help us find key areas of expertise and work experience when conducting professional searches.

All of these things are examples of efficiency improvements – we can and did operate without AI when doing these tasks previously but this technology can speed things up. That is, in my view, the right way to approach using AI given its present capabilities, though.

Right now, there are many examples of people using AI ineffectively and, ironically, inefficiently. This can mean asking for tools that don’t add value or that even already exist, so collectively we must temper our expectations.

While it’s easy to understand why some people are anxious about its impact on the global workforce, the idea of replacing entire swathes of the population with AI is just unrealistic at this stage. Those fears about its impact on employment shouldn’t be ignored and while regulatory efforts are catching up, any legislation and controls must be fit for purpose and futureproofed as much as possible.

The reality is almost nobody knows exactly what will come next, whether that be good or bad, but this technology is not going to go away so we need to understand how best to interact with us. The technology, society’s understanding of how best to use it, and control systems improve, we should be able to take bigger, bolder steps in how we apply it.

For now, though, many of us just need to ask ourselves this about using AI in a work context: does using this tool actually help me and does it speed up my work or make life easier? When they do, they will realise they themselves will be able to at least as good a job as this tool for many tasks. They will realise this technology is a means to an end, rather than an end in itself. They will realise is that all we really need is a sprinkling of AI.