Why prompt engineers could be the next big role in agencies

Everyone in our industry is staring intensely through the AI lens now, weighing up the positives and negatives. But regardless of which camp you sit in, there's no denying it will be a big bang moment – with a monumental impact on jobs akin to the last industrial revolution.

While AI will automate some jobs, a PwC AI study has found that "Any job losses from automation are likely to be broadly offset in the long run by new jobs created as a result of the larger and wealthier economy made possible by these new technologies."

But even if the jobs lost and created roughly balance each other out, there's still the issue of who will do these new jobs.

The World Economic Forum has said that 50% of all employees will need reskilling by 2025 as the adoption of technology increases.

They say you need to do something for 10,000 hours before you become an expert. So, all you're doing is learning about Generative AI for 10 hours daily.

That's two years, eight months, and 27 days. So, we don't have many experts out there just yet!

'Prompt engineering' is communicating effectively with an AI interface to achieve the desired results. As AI technology continues to develop at pace, being able to master prompt engineering is going to become a sought-after skill.

The default in many business situations when new technology comes along is:

Step 1. Block it.

Step 2. Realise its importance but reintroduce a sanitised or heavily policed variant later.

Step 3. Fall behind your competitors and end up playing catch-up forever more.

I understand this rationale from a legal, IT, security and data protection point of view. But it instantly puts these businesses on the back foot. That's where the role of agencies comes in. Those businesses will be reaching out to their agency partners as the experts.

Remember those 10,000 hours? If you aren't chipping into that now, you should be.

For example, we're currently testing generative AI to help produce regular content at scale for selected clients. It's a steep learning curve to be scaled quickly, but it's reaping immediate benefits for them and us.

Here's the thing when it comes to the most innovative technology...

Take a modern camera. Anyone can pick one up, press the shutter and get a more or less decent photo. You don't need to be Rankin or Annie Leibovitz.

But if you take some time to study all the functions and settings, you'll take an even better photo. Fact.

Next is the most unquantifiable thing. Creativity.

Having an eye for a good photo or composition isn't something you can expect AI to know. It isn't sentient. It's not human. AI doesn't have an imagination. It needs to be fed with the correct information to create something great. You must be able to describe it in as much detail as possible to get the best results back.

To do that, you need to think like and speak the language of a photographer, illustrator, screenwriter, author, web developer, coder, etc.

I've watched hundreds (no, really) of videos on YouTube of the "I built a website entirely in ChatGPT" variety. But every time I've seen one, it's clearly from an individual with a deep knowledge and understanding of HTML5, JavaScript and CSS. An expert who's able to piece together the elements ChatGPT goes on to generate. And, because they know the right things to ask the AI model, they get the best results.

I can say with absolute certainty that if I dared ask my Mum to build a website using ChatGPT, she'd probably exclaim, "Web-what?" and gladly return to watching Strictly. (Sorry, Mum.)

So, while developers and creatives will remain in high demand, those who possess a deep understanding of AI tools and can skillfully engineer them with the appropriate prompts will be immensely sought after.

The other situation I see arising is very specialist prompt engineers with a more profound knowledge of the AI algorithms and platforms they're dealing with. They can work with creatives to generate and then direct the engineer to make the changes and adapts to suit.

Regarding web development, it's akin to having specialists within the field. You might come across job postings like "Wanted - An Umbraco, MySQL, Javascript, WordPress, XXXXXXX developer to work on X project”.

I can see a similar situation with AI – "Wanted a mid-journey, Leonardo, ChatGPT, XXXXX Prompt Engineer to work on X project".

One thing is abundantly clear, the landscape as we know it will change, not in the next decade. It's already started and is changing month by month.

ChatGPT only launched in November 2022, and it's already the 21st most visited website in the world.

Generative AI is here to stay. Whether we like it or not, it's up to you to decide if you want to jump in at the deep end or bury your head in the sand. I know what I'm doing.

LOOP Agencies are already working with AI Tools to provide a more efficient service to our clients to generate content; if you want to know more, get in touch.