Cannes the Bots win?
Bots will make award winning work. Plus Tipsy, a new bot to help you win at Cannes.
A few weeks back I was invited by the good folk at Contagious to speak at FutureFlash — the conference put on by the Canadian Institute of Communications Agencies. I mostly talked about how brands might work in a world where consumers all have chips embedded in their brains. (BMI’s, Elon Musk’s Neuralink. I’ll do a post on that soon). But one slide was about the current state of creativity and bots. It started a little debate so I thought I’d expand a little.
But before that. If you are visiting Cannes this year, check out a bot we made with Huge. Tipsy is a bot that helps you make the most of your week. Tipsy gives you tips on how to find things, nice food, cheap(er) drinks and where the parties are. Tipsy also has a really neat translate feature that allows you to type in anything and…have it translated.
As we always say, the best bots are both entertaining. Try out Tipsy here.
Ok, back to the prediction. In the very near future — maybe even this year — there will be a piece of creative work that is made by bots that will win an award at Cannes.
Some clarification. I don’t mean a campaign bot for a brand I mean an algorithm actually making the work, making the creative decisions. Similar to the trailer that IBM Watson did for a movie. I guess they’ll probably enter that to Cannes so who know’s maybe that will win?
The reason that an AI was able to make a horror movie trailer was that they are very formulaic. The IBM team fed in hundreds of classic horror trailers and Watson was able to — actually quite easily — figure out the patterns; for example a series tense scenes followed by quick cuts of dramatic scenes.
It’s all about input.
I have always believed that the best advertising creatives are simply the people who are able to make the best connections. Not in a networking sense, but in a cultural sense. Seeing movies, reading books, going to galleries, museums, talks, taking pictures, traveling the world, making notes. You have to be able to conjure up this stuff when the time (the brief) is right and then apply them in an unusual way. That’s what creates the feelz. When done well it seems incredibly, almost unfairly, simple. What if we made a bronze statue of a little girl and put it next to the bull on Wall St? What if we took a weird techno song and made a black and white surf movie with a horse in it. The better input you have, the better your output.
If there is one thing that computers are good at it is collecting and categorizing input. A bot would do this very well. For sure there would need to be some human input at the start to figure out what the categories might be, to determine what is a ‘good’ or ‘worthwhile’ input, but that seems relatively straightforward. For now, at least, all bots have a human creator at the beginning.
The IMB horror movie was finally edited by a human. A human also made the music choice. You could argue that without this human touch it would not have worked as well but that misses the point. In a few days I can teach a bot what music works well by looking at other music choices and how to edit even better by analyzing great editors.
Bots like formulas. So do awards juries.
Horror movie trailers are very formulaic. But actually winning awards is also very formulaic. It’s just a different kind of formula. It doesn’t appear to be so obvious as a horror movie trailer. In fact the formula for winning at Cannes is to be different — but not so different enough that no one can understand what you’re trying to achieve.
I used to be on the global creative council at a big network ad agency. One part of my job was to look at work from around the network and figure out how it could win points at Cannes. Using nothing more than past winners and gut instinct we would try to push things in a certain direction. And in many cases — you’ll be shocked, truly shocked! to learn — we changed work to have a better chance of winning. A bot could look at work that was bubbling under from last year but didn’t quite cut through to establish some potential trends that might hit this year. A bot would do that better than a human.
There was also a very hard working team of people who would look at all the categories and figure out where there were gaps and potential places to pick up points. I’m sure they would have rather have had a robot do that for them.
Remember everything that the bots are doing here can be done endlessly. It costs the same for the bot to do this for one Cannes entry as a million entries.
Is this the end of Creative Directors then?
As explained there are two parts of the process of winning at Cannes. The initial creation and the strategy for entering. Both of these could be done very efficiently by a bot. Does that mean bots will replace Creatives or Creative Directors? No I don’t think so at all. One of the things that the bots will figure out is that after a while fashion becomes a fad. For one year everyone will say, ‘oh cool, that film or print ad or whatever was made a by a bot. Clever’, but the year after that it will feel outdated. The humans will have to come up with something else. Humans are good at that.
Dexter is our platform that helps anyone build a bot. It is the lovechild of Squarespace and Mailchimp. If you can write you can build a bot on Dexter. We’ve had Drake Bots, Fuckboi bots, Dad Joke bots — all built by people who don’t know how to code. So what you waiting for? Try it out.
Tipsy was also built on Dexter. Check it out here.
