In AI World, Pattern Disruption Trumps Pattern Matching

Ali Abdaal fits a certain profile.

He’s a “doctor turned YouTuber,” who tweets about “productivity, personal development and life as a creatorpreneur.” He’s got a newsletter called “Sunday Snippets,” where he shares “actionable productivity tips and practical life advice.” He’s got more than 3 million followers on YouTube and -- of course -- hosts a course on how to grow your own YouTube following. On Twitter, he refers to himself as a “thinkboi.”

Earlier this week, he had a thread go viral, generating one of his highest-performing tweets of all time. The topic was what you’d expect: 15 actionable productivity tips. There was only one issue: He didn’t write it.

In fact, nobody did -- no human, anyway. It was written by an AI system called Lex, following a few prompts he supplied. Abdaal detailed the process in a follow-up thread.



This kind of story is by no means isolated. AI-enabled (or even fully generated) copywriting is becoming more and more common. Some of it is shockingly good; all of it will only get better.

Stephen Marche makes this observation in an essay for The Atlantic called “Of God And Machines,” noting: “And if AI harnesses the power promised by quantum computing, everything I’m describing here would be the first dulcet breezes of a hurricane. Ersatz humans are going to be one of the least interesting aspects of the new technology. This is not an inhuman intelligence but an inhuman capacity for digital intelligence. An artificial general intelligence will probably look more like a whole series of exponentially improving tools than a single thing.

"It will be a whole series of increasingly powerful and semi-invisible assistants, a whole series of increasingly powerful and semi-invisible surveillance states, a whole series of increasingly powerful and semi-invisible weapons systems. The world would change; we shouldn’t expect it to change in any kind of way that you would recognize.”

He follows it up with: “The paragraph above was composed by GPT-3. I wrote up to ‘And if AI harnesses the power promised by quantum computing’; machines did the rest.”

I would love to be able to write that beautifully.

It's not just copy, of course. AI-generated art is entering mainstream capability with tools like DALL-E, Stable Diffusion, and MidJourney. And the obvious next step from text-to-still-image capability is text-to-moving-image capability, with tools from Runway and Meta.

But back to Abdaal's viral post. That an AI was able to generate his tweet thread shouldn’t be surprising to anyone following the space. It followed a predictable pattern, a standard formula. It’s in listicle format, which doesn’t require original insights or bold new ideas.

For now, the post was different enough that it works. But this format, like every templated form of creativity, is rapidly becoming oversaturated. And with AI accelerating the diffusion of the template, we’re likely to reach saturation that much sooner.

The marketing skill of the future, therefore, is not to understand the tried-and-true patterns that evoke intrigue in your readers. It is to understand that your readers will rapidly grow to assume any writing that follows a tried-and-true pattern was probably generated by a machine.

Our jobs, therefore, as media people, as copywriters, as marketers, become less about leveraging well-understood patterns and more about disrupting those patterns. And we can’t just disrupt randomly; we have to do it in such a way as to be seen, heard, appreciated, in such a way as to still drive engagement and emotional response.

Does that make our jobs harder than they’ve ever been? Yes, it does. But while we’re railing against it, the AI is improving all the time -- so we’d better start practicing now.

1 comment about "In AI World, Pattern Disruption Trumps Pattern Matching".
Check to receive email when comments are posted.
  1. James Smith from J. R. Smith Group, October 22, 2022 at 12:07 a.m.

    While there are stellar examples of great AI output, some AI specialists remind us that most AI applications are only as good as the material they are trained on and/or continue to learn from. As such, a human "wrangler" is still required to oversee and constantly monitor inputs and outputs.

    AI is getting better at routine tasks, such as writing basic email copy, but you are right to stress that successful marketing is often about disrupting known or normal patterns, which points to a classic Leo Burnett quote.

    "The secret of all effective originality in advertising is not the creation of new and tricky words and pictures, but one of putting familiar words and pictures into new relationships."

Next story loading loading..