I was wrong about AI, marketing & creativity
thoughts on my generative AI skepticism two years on
Two years ago, I posted this thought on LinkedIn.
Turns out, I think I was wrong, and totally missing the forest for the trees.
It’s obvious to anyone watching, that AI is going to be, and already is, a very good creative sidekick. Yes, you still need to have something good to say (ideas and insights still matter), but my assertion that an hour spent with a whiteboard is better than an hour prompting, feels just plain wrong to me now.
Because the truth is, prompting with AI is a kind of whiteboarding. It's messy. It’s brainstorming. It's fast. It can surprise you. It can take you down the wrong path, but also unexpected paths.
I get it though. There is a romance attached to the idea that the best ideas should just appear out of thin air, maybe on walk in nature, or in the shower, or if you just put enough creative people in a room. A sort of “slow food” or artisinal approach to ideas, innovation and creativity.
That technology somehow taints the organic natural beauty of creativity.
But at this point, I honestly wonder if you can be as creative without AI.
Maybe that’s too hyperbolic, maybe my POV pendulum has swung too far the other way, but I think the gaping hole in my logic back then was that I, like many, was thinking of AI as a factory worker: linear, derivative, mechanical, soulless.
But I, like many, now understand AI isn’t that at all. It can be fun, creative, random, energising. It’s a force multiplier if applied well.
It was kinda unclear in early 2023 to be fair (or at least it was to me), but there is now nothing to say AI cannot come up with as cool, as crazy, as divisive, as rebellious ideas as a great creative.
A bunch of the world’s top agencies have started to invest in initiatives to build “creativity benchmarks” for different LLMs and there is increasing evidence that just as we’ve trained LLMs to be better at maths, coding or writing, we can train them to be more “creative” too.
I guess the other thing I didn’t see back then was that, AI would become more than just a chatbot and text. More than just a playground for ideas. I didn’t quite see that the multi-modal nature of generative AI would change everything, and that the time-cost-and-effort-gap between ideas and execution would shrink drastically. And what that would mean for marketing, creativity and experimentation.
Veo3, Google’s text-to-video model is frankly mind-blowing and already producing both interesting art and marketing. A brand recently aired an ad made with AI during the NBA playoffs. Apparently it cost $2,000 and took “just days” to make.
The other day, a “fake ad” for the internet-darling brand Liquid Death, made by Google did the rounds online. As always, it divided opinion. To be clear, the video itself is produced with AI, but the script was not made by AI, according to its maker Amir Ariely.
This ad, apart from being absolutely hilarious and very impressive technically (the cuts don’t feel too janky, characters feel consistent etc), tests well too. The ad testing company System1 tested the ad, and say “it beats 75% of water ads they’ve ever researched/tested”.
An important detail is that, apparently the fake Liquid Death ad cost just $800 in veo3 credits to make. Meanwhile, the “old world” brags about how much time and effort it takes to get a 60-second commercial out.
Don’t get me wrong, the Nike ad is probably great. And sometimes great work/art can take time. But the Liquid Death ad is pretty great too. And there’s no inherent or intrinsic value in something taking a long time. It only really matters if it’s good, whether it took 6 days or 6 months is irrelevant.
Just like it doesn’t matter to us if the song we like was made in a bedroom or in an LA studio.
And my evolved view now is that AI can help us make good things, better, faster and cheaper. To be sure, it’ll produce a lot of crap too. But that’s beside the point.
As investor and former Adobe product executive has said, AI will both lower the floor, and raise the ceiling for creativity, at the same time.
So yeah, I was wrong about AI and creativity. I now believe it can make us both more productive, and more creative. Not to mention make things faster and cheaper.
That’s truly exciting for creatives and marketers.
I wonder where we’ll be in two years from now.