WARC recently published this article by our Senior Creative Technologist in London, Sam Cox analysing the current use of AI within the marketing industry and his hopes for how it might shift into a force for real change in the future. They've been kind enough to let us reproduce it below.
Artificial intelligence is out of control. But not in the Blade Runner, Terminator, lock-up-your-loved-ones kind of way – it’s out of control in the far less terrifying world of marketing.
AI is marketers’ current obsession – at Dmexco this year, 57% of 186 participants surveyed said they saw AI as the most important trend at the conference. Everyone is talking about it. As such, it’s in danger of becoming the latest in a long line of marketing buzzwords, something that’s overhyped and used for the sake of being seen as innovative.
AI has been on an interesting journey over the past decade. It’s risen from relative obscurity within the public domain to a key subject at every tech conference around the world. The first time most marketers really started to pay attention to AI’s potential was when chatbots became the Next Big Thing, circa 2016. Everyone wanted to be perceived as an expert in bots and this led to them being produced en masse, often without any real strategic thinking as to the value they would provide to the customer. While some proved useful and continue to be used, many of them now lie dormant. The rise and (partial) fall of bots should teach us some lessons about the next evolution of AI.
We should never try and use a particular technology as the first port of call. Good marketing starts with a problem to be solved, and then identifying what technology works best for that particular solution. This might involve AI, or it might not. Concocting an excuse to use AI will lead to a far higher likelihood of your product failing.
That’s not to say we shouldn’t be experimenting with new technologies – it’s hugely important from a learning, development and future-proofing perspective to be up to speed. Here at We Are Social, we’re currently working on proactive client projects that involve AI, where it’s the best fit for the problem in hand, and we also experiment internally.
You need to understand technology in order to know when best to apply it. There will come a time when the fit for the latest, most innovative solution is right. But this isn’t something that’s going to come from a 30-minute brainstorm; it’s a long-tail effect and a long-term investment. True innovation and groundbreaking products and services take longer than a couple of days to make.
I’ve seen a few great examples of AI recently. Adobe’s AI and machine learning framework, Adobe Sensei, is able to design a film poster from a sketch to a layout in just a few minutes, while creating versions for targeting different groups. While by no means eliminating the need for human input, you can see how this could make the creative process more efficient.
IBM’s AI-driven automation improved customer experience and storytelling at Wimbledon this year. It helped an editorial team build creative highlight reels based on the most remarkable moments of play, analysed by players’ facial expressions. TeleportHQ has built real-time code generation using computer vision, which builds code as a person sketches wireframes on a whiteboard.
It’s certainly worth noting that all of these examples make a person’s job easier, rather than actually replacing said person. There is always room for the human trait of creativity – this will be one of the last skills for AI to master.
With 2018 rapidly coming to a close, I would love to see 2019 as the year that marketers stopped thinking about AI as the starting point, and more about how to use technology in a way that makes people’s lives easier. We’re lucky to live in a time when the pace of technological advancement is incredible.
Keep up – but don’t get caught in the hype.