ChatGPT AI copywriter: Looming threat or opportunity to evolve?
May 4, 2023 | AI, ChatGPT, artificial intelligence
Even as someone who makes a living working in healthcare communications, I can still spend many hours copywriting without much progress – it is the nature of the beast [come back to this bit later and write something brilliant].
So, to hear that ChatGPT – a cutting-edge chat bot built upon some seriously clever language-based artificial intelligence (AI) – is effortlessly producing copy, made me feel uneasy. I reassured myself that it wouldn’t be able to produce anything of any true quality and darted over to check it out. I was disturbed to find that with minimal effort on my part, ChatGPT was able to provide me with copy so strong, that I considered outsourcing my weeks’ to-do-list.
Naturally, people are often resistant to change, but I would never think of myself as one of those people. This change however, felt like it had the potential to make my skill set irrelevant. Technology does that sometimes. There are certainly fewer cashiers than there used to be in my local supermarket, and that point is illustrated further by looking at the 84,000 people employed at Blockbuster Video during its peak, versus just 11,300 currently working for Netflix. Technological unemployment, as this is known, has been discussed by economists for many years as they try to size up what threat technology poses to the global economy. Whilst it is accepted that technological advances can lead to short-term job-loss, opinions are split when looking at the long-term picture. While some economists believe that there will always be enough new jobs created by technology to compensate for those jobs lost, other economists do not. If the world’s best economists are torn, then that is certainly not a debate I will be entering. What interests me, from a selfish perspective, is the consensus that short-term job-loss will occur. Not only do I create (scientific) copy, but as an agency, Page & Page and Partners sell it. It makes up a significant percentage of our revenue. So, we need to give ChatGPT and everything it represents some serious consideration. The question is, how do we approach technological advances that might present a threat to our value proposition, and, potentially more importantly, how can we adapt?
A quick peruse of my LinkedIn feed (representing a biased sample of agency professionals) shows some thematic responses to ChatGPT, forming three distinct groups of people, in my mind at least.
The deniers
This first group is often quick to point out all the shortcomings in the technology and how far away it is from replacing a human. Yes, ChatGPT still has accuracy problems, and it is yet to ‘understand’ a request – copy is more of a reflex than a thought-out response. But, are all technological advances not based on iterative improvements on early models? Surely, ChatGPT is just in its ‘early model’ phase, primed for such improvements. Will humans always hold the monopoly on insight and nuance? I understand the relevance of these opinions, but I hope it is paired with more opportunistic thinking. Otherwise, I fear this group are the brass band playing whilst the Titanic went under.
The opportunists
Across the Atlantic, in Princetown New Jersey, Edward Tian, a 22-year-old Princetown student had been studying how to detect text that had been created by AI long before ChatGPT arrived. And, when he witnessed first-hand the potential power of ChatGPT, something stirred inside of him. It wasn’t just the idea of jobs being at risk, but also concerns over the authenticity of the produced text. How will we know if something has been created by a human or a machine? Tian had the skills and expertise needed to create the antidote to AI-generated copy; GPTzero (gptzero.me ). Almost serving as a mirror onto itself, GPTzero is an AI-powered detector of AI with the simple but equally pertinent tagline ‘humans deserve to know the truth’. Since its creation, GPTzero has ‘gone viral’ – it has been featured by pretty much every big news channel and received approaches from plagiarism detection services like Turnitin. Opportunists like Tian are building completely new services or products in response to these rapidly developing technological advancements.
The adopters
Adopters are those that will integrate the technology into what they already do, evolving with it. I am sure that there are many individuals, that fit this mould – certainly students with looming deadlines and an unfinished essay might be likely to employ some AI assistance if it is likely to go unnoticed by their stringent approval systems. In the corporate world, digital content producers seem to be the first movers. Buzzfeed, for example, have paired up with OpenAI (the creators of ChatGPT) to automate the creation and personalisation of their digital content. In a move that was eerily close to this pairing, Buzzfeed also announced late in 2022 that they would be laying off approximately 12% of their staff, having felt the strain of the deteriorating global economic climate. Whilst correlation does not always equal causation, perhaps this is technological unemployment in real-time. Whilst these harsh realities and rapid changes may be a difficult pill to swallow for some, the leaders of tomorrow will be defined by how quickly they adapt today.
The implications for agencies
For creative communications agencies, it is not just copy that is becoming automated. Midjourney AI is a powerful tool for artistic image creation, and just one example of AI being used to create images and advertisements. For many designers, illustrators, and art directors, this elicits the same anxious response that ChatGPT elicits from copywriters. But image generation AI also has shortcomings. The outputs are highly unpredictable, and highly dependent on the user’s ability to ‘speak the language’ of the software when ‘briefing’ the software.
Under threat from all angles, what is our next move? I asked a famous polymath, an expert in the field of AI and one who has studied over 8 million articles related to the subject.
Page & Page: AI poses a potential threat to creative agencies, how can these agencies change to remain relevant?
ChatGPT: While AI can help in the creative process, it can only go so far. AI can pull from big data to create trends and give insights, but it does not have human understanding at its core. This is where creative agencies can live up to their name and inject their creativity into the mix. There is always a need for human minds in the areas of focus groups, advisory boards, brainstorming sessions, and research in order to be able to ask the right questions that work best with AI.
That sounds encouraging, but then again, you would that say if you were trying to steal our jobs…
I think we can all agree that those who just deny the impact that AI is having, and will continue to have on our profession are the laggards of our technologically-focused world, and are simply doomed to fail. The beneficiaries of change are those that are always on the look-out for opportunity and are willing and able to adapt accordingly. For companies, the idea of Tarzan economics is just that. Knowing when to let go of the old vine and grab onto the new vine to carry you forward. Timing is everything – hold on too long and you may soon find yourself moving backwards.
The question we must ask ourselves, is where will this new vine be leading us, and what might the next vine involve?
By Kristian Webb & Emily Sidlow, Page & Page and Partners
This content was provided by Page & Page and Partners