Aurora Magazine

Promoting excellence in advertising

Your Very Own Superpower Ingredient

Julian Saunders on what the coming AI wave could mean to advertising practitioners.
Published 13 Mar, 2024 11:24am

It is not just a tool or platform but a transformative Meta-technology, the technology behind technology and everything else. Itself a maker of tools and platforms, not just a system but a generator of systems of any and all kinds.

In his much-awarded new book, The Coming Wave, Mustafa Suleyman defines why it is so difficult to predict the impacts of AI and why they will be so pervasive. It is “a general use technology” – a superpower ingredient that will find many applications. Some predictable, some not yet imagined.

Here is a thought experiment for you about an earlier “general use technology.” Imagine you were alive in the 1880s. Thomas Edison had invented a way to turn electricity into light – the light bulb. If you had engaged in futurology back then, you would probably not have imagined the following. The movies, factories that could work in 24-hour shifts, and the multiple ways that humans would be liberated from the dark to live out their days in radically different patterns no longer limited by daylight hours. And much more.

Arguably, AI is moving into the mainstream as a pervasive technology much quicker than the electric light bulb. It is already integrated into our daily lives. It has already become unremarkable and taken for granted. Here are some examples you may have already experienced:

If you have an X-ray, it is probably being analysed by an AI system which is better at spotting health conditions than a human being.

Virtual assistants like Apple’s Siri, Amazon’s Alexa, Google Assistant, and others, use AI algorithms to understand and respond to user commands and queries.

Recommendation systems are found in online platforms like streaming services (Netflix, Spotify), e-commerce websites (Amazon, eBay), and social media (Facebook, Instagram). These systems analyse user behaviour, preferences, past interactions and similar user profiles to suggest personalised content, products, movies, music or connections.

Predictive text and auto-correction features are commonly found in messaging apps and word-processing software. They use natural language processing and machine learning to predict the next word or phrase you may intend to type.

ChatGPT was developed by a well-funded company, but this type of large language model (LLM) will become cheaper in quick time. Remember when a website could only be afforded by big companies in the early 2000s? A decade on, anyone could put together a professional-looking e-commerce site using pre-formatted elements for a few hundred dollars a year. A similar dynamic is at play as different industries develop AI systems designed for their specific needs.

The Chinese Communist Party is using AI in its surveillance systems, especially visual and gait recognition. Mindful of the strength of US tech companies, China is committed to becoming a leader in AI, which is unleashing a parallel wave of innovation in areas such as e-commerce and autonomous driving. Huge resources are being committed to AI-powered innovation and this will flow through into applications in many sectors, through a combination of geopolitical competition, state-funded innovation (especially in universities), and private enterprise.

In his seminal book, How Innovation Works, Matt Ridley said this about the development of the light bulb: “The truth is that the story of the light bulb, far from illustrating the importance of the heroic inventor, turns out to tell the opposite story, of innovation as a gradual, incremental, collective process yet inescapably inevitable process. The light bulb emerged inexorably from the combined technologies of the day.”

You could substitute the words ‘artificial intelligence’ for ‘light bulb’ in that last paragraph to understand its innovation dynamic. Digitally stored data, the huge processing power of modern computing and cloud storage are among the combined technologies that enable the development of AI.

In the next few years, you can expect many invitations to conferences about AI and plenty of dystopian (‘many jobs are going to disappear’) and/or utopian predictions (‘more rapid development of cures for diseases’).

AI will certainly change the workplace. New roles will emerge and others will decline, especially in advertising and marketing. Twenty years ago, for example, search engine optimisation, search engine marketing and social media marketing did not exist. Meanwhile, a whole craft industry in photographic retouching of physical negatives was wiped out by our ability ‘to Photoshop’ digital images. In creative departments, Midjourney is already being widely used to generate images from verbal prompts, which is putting artists and visualisers out of work. However, a new skill emerges – how to write clever prompts to develop distinctive and usable images.

Response-type advertising will increasingly be automated. It has long been a candidate for automation as its techniques are well-known and follow a pattern. The likes of Meta and LinkedIn will offer advertisers a service covering creative development, creative execution and media planning. These will be easy-to-use self-service platforms that will also optimise your campaigns as the system learns what works for your particular offer. It will be liberating for those who may take their advertising in-house. Ad agencies will have to adapt.

In my training courses, we use ChatGPT and Bard (Google’s version) to develop creative strategies. Students are quick to understand how to acquire knowledge on a whole range of topics; trends/opportunities in the market, the brand, its competitors and who the likely customers are. (Be careful though, because ChatGPT’s data has not been updated since January 2022 and will not be current when asked a question like ‘What are the emerging trends?’ in a particular category).

Some of my students are going further and using AI as a tool for creative inspiration. They are using prompts to get ChatGPT to write scripts or brand positioning, for example, “in the style of Apple, a rapper, or Dylan Thomas.” The best students take time to experiment, run down rabbit holes and play. The platform may not give the final answer but it could deliver a first draft quickly or uncover a fresh angle. This way of using AI is already an emerging skill among creative types. To do it well you have to bring your own imagination to the party.

My advice then is start to experiment. The only cost will be your time, as ChatGPT 3.5 is free and the ‘plus’ version has a $20 a month subscription. As the platform trains itself with more data, it will only get better. We have all learned new behaviours in recent years such as how to Google, how to be influential on social media or how to run our affairs using a smartphone app. Learning how to write productive prompts on AI platforms comes next.

I doubt AI will replace human beings but it will be a productivity and creative tool that smart marketers and agency folk will increasingly use.

Julian Saunders was CEO of Redcell Advertising and worked in an innovation team at Google.