A.I. is not going to be what anyone thinks it will be. Yet from a marketing perspective1 it seems like we’re confident we’ve figured it all out.
We’ll use generative A.I. to split test a bajillion versions of a thing. We’ll embrace whatever crumbs the generative engines drop on how their algorithms decide what to reference and surface. We’ll promote whatever new way our company has infused A.I. into its products and services, regardless of whether it’s aligned to established personas or brand guidelines. And we’ll do it all based on how A.I. is being used at this exact moment.
The fact is, people (including your target customers) are still figuring out how they’re going to use A.I.
History tells us that sometimes it’s just a matter of becoming comfortable with it. In the 90s, a lot of people couldn’t fathom that it was going to be o.k. for them to input their credit card information into Netscape as they surfed the World Wide Web. In that case, technology and security and user behavior evolved together.
Sometimes people are the ones who drive the change. Flickr started as an online game which happened to have a photo-sharing feature. The people didn’t care about the game, but they liked sharing photos. The Twitter platform integrated functionality around hashtags and retweets because people were already retweeting and hashtagging things on their own.
This happens all the time. A new technology is typically developed in a way that aligns with human behavior based on current behavioral information,2 and with an expectation of that being how it will be used. But using that new technology actually changes human behavior as we learn to interact with it. The technology is then adjusted to reflect that new behavior, and then in order to fit the updated technology the behavior evolves again, and then so too does the technology — forever and ever amen.
Once when I was driving into work I was about to change lanes but there was a car there. There was maybe enough room for me, but only just, so I wasn’t going to risk it. But then I realized the car was a Tesla, and that it would slow down if I pulled right in front of it.3 New technology was impacting my behavior.
Knowing this happens is nothing new. In the 1962 book Diffusion of Innovations, Everett Rogers wrote, “Until very recently we assumed that adoption of an innovation meant the exact copying or imitation of how the innovation had been used previously in a different setting. Sometimes the adoption of an innovation does indeed represent Identical behavior … In many other cases, however, an innovation is not invariant as it diffuses.”
New technology will simultaneously reflect, change, and adapt to human behavior. Which is why it amazes me how definitive so many people, and marketers specifically, are being about A.I. and its integration with the discipline. Extrapolate all you want, but A.I.’s future, its potential, its applications, and its ultimate use are going to change dramatically as we go, and any foundations you build right now are probably going to need to be demolished at some point.
So how about we take a deep breath and be intentional about what exactly we’re trying to accomplish, and whether the newfangled whatever really is the best way to accomplish it? That’s the approach that will create lasting value while everyone else is preoccupied chasing the shiny new thing.
1 I’m not even going to get into all the A.I.-based startups and SaaS solutions and whatever else that have popped up.
2 Unless you’re Apple in the ‘00s when Steve Jobs just kinda did whatever he wanted with iPod click wheels and whatever else and consumers followed suit.
3 I, of course, didn’t do that; but a lesser man might have.