Rothko wrestles with the AI revolution
An estimated 30% of jobs are likely to be replaced by generative A.I. – as early as the mid-2030s. With it comes the increased threat of disinformation campaigns and the (already realised) possibility of deep fakes making mass manipulation that much easier. The recent uncommissioned viral hit song mimicking the style of rapper/singer Drake and The Weekend gave the world a taste of how generative A.I. can digest an artists’ 17-year body of work at warp speed. The result was a believable imitation that the artists’ music label found to be a most insincere form of flattery. This article examines the threats to copyright and authenticity, as well as the possibilities to be leveraged to keep up with trends in the industry.
By Hailey Gaunt, Writer, Rothko Brand Partners
A.I.: friend or foe?
As a brand and public relations agency, words and design are our currency – we know we can’t afford to press snooze and hope to wake up once benefits of adopting this technology have been exploited by others. Some of us are tech-savvy, early adopters – naturally curious and enthusiastic about the prospect of expanding and expediting what we do. These folks are confident in their ability to adapt. Others are far more tentative – anxious about how the hacks will cheapen the product and rob us of opportunities to apply our carefully-honed skills gainfully. All of us are wondering what the tech infiltration will mean for the business of writing and design if clients can use these tools rather than the services of a professional.
We know, however, that treating A.I. as the enemy would be unwise. And perhaps in the spirit of ‘keeping your friends close and enemies closer’, it’s a good idea to get comfortable with the competition.
Experimenting
We tried a little exercise at a recent writers’ workshop with our in-house team of writers where we gave each a mock brief from a client – a non-profit organisation working to combat ocean plastic pollution in Cape Town – asking them to prompt ChatGPT or Bing to create an op-ed with a compelling headline and lead. They had five minutes to try and refine their results. Most results were generic – basic statements about pollution without a relevant local angle, decent hook, or evidence to substantiate the claims – certainly no unique voice. One output was amusingly poetic: “As the sun sets over the breath-taking shores of Cape Town, an invisible menace lurks beneath the surface...”
With a bit more time and instruction, the bots might have come up with something better. But the basic conclusion we drew from the fifteen or so A.I.-created op-eds was that even if the bot had bulked up its offering, a piece of writing is far more complex. What makes a compelling read is the voice behind it – the author whose own ideas and hard-won opinion must convince, persuade and guide a reader through their mind. It’s not just the what, but the why of a piece that makes it effective and, lacking sentience and real-world experience, the bots can’t provide the why.
Will ChatGPT’s ability to competently construct words pose a threat to much of the feckless writing, full of borrowed ideas and repurposed content? Absolutely. But from the standpoint of these early days, we believe that the powerful proliferation of words and designs will only put higher value on good words and excellent arguments.
In the case of thought leadership, op-eds and the like, we will have to work more closely with our client-experts to help them identify what their most salient, relevant knowledge is, and how to link it to topical issues that can give it footing in the public sphere. We’ll have to be savvier about getting these ideas into the world in the right publications, for the appropriate audience. We’ll have to do more to make the voice of clients ring sharp and true against the maddening crowd.
From a client-care perspective, we need to be more deliberate about our relationships – getting to know who they are, how they think, intuiting their needs – proving the value of real relationships cultivated over time.
So how can we use A.I.?
Let’s get practical. We’re interested in how we can use the tools at our disposal to make what we do even better. We’re not talking efficiency for its own sake but enhancing our understanding and refining our offering.
We asked our writers if and how they’d been using large language models and were surprised to learn that they had started wading into the waters, using ChatGPT, for example, to help explain the intricacies of economic concepts or to find the flaws in an argument. Many found it good at digesting jargon or refining language to suit a target audience. Some were using it to uncover a fresh angle on a topic – identifying what had not yet been written.
For writers working with veteran thought leaders, the curation abilities of Bing helped them to easily find and absorb an author’s voice, based on what they’d already published.
As an editor, forever in pursuit of compelling headlines and section breaks to sign-post to readers, I have also had fun (and some success) prompting Bing to help enhance these.
Getting real
As a writer, editor and creative who has spent my adult life trying to come up with words that matter, I’m not thrilled by the implications of on-demand text generation. I became a writer not because it was an easy, automatic process to a thoughtful, coherent – dare I hope – beautiful, piece of writing, but because it is in fact not. For some reason I’ve found this particular struggle satisfying. When I came across musician and multi-disciplinary artist Nick Cave’s response to a fan letter attempting to impress him with “lyrics in the style of Nick Cave”, I circulated it widely and felt vindicated to be on the side of a creative genius. Cave captures his sense of doom most poetically, arguing that ChatGPT could never write a genuine song because songs and art arise from suffering and lacking “an authentic human experience”, forcing A.I. to mimic what already exists. “ChatGPT’s melancholy role is that it is destined to imitate and can never have an authentic human experience, no matter how devalued and inconsequential the human experience may in time become.”
At Rothko, we still think that human experience and connection hold value and, in fact, underpin our service – but we’re keeping a close eye on the competition.
You may also be interested in
