When Philippe Warzée asked me to give my point of view on the use of A.I. in my daily life, I said to myself: “Nothing could be easier! That’s obviously presumptuous, even though it’s been 5 years (after a solid theoretical introduction by Professor Hugues Bersini, who has devoted almost 40 years of his teaching career to artificial intelligence at ULB) since I had the chance to share a summary of the role of AI in marketing and communication with the students of the Executive Master in Digital Marketing at the Solvay Brussels School of Economics … in theory, then, I shouldn’t have had any trouble finding an engaging angle. But what about in practice?
Original source: Il faut en effet remonter à “Second Life” (2003) pour trouver une révolution technologique qui génère autant de réactions divergentes | Room2902 (pub.be)
One indicator is revealing: the news of the last few weeks (in other words, the widespread availability of ‘open’ or ‘generative’ solutions) has meant that I have reworked the content of my course more often in 4 months than in 4 years. After an introduction on the acceptance of technologies by consumers, the plan of this course is built around 3 themes: What are the opportunities for exploiting A.I. in the ‘Marketing Mix’, how has A.I. changed the ‘Consumer Decision Journey’ and what will be the consequences for employment in the world of communication and marketing.
However, the introductory theme of this course has remained unchanged: the new “Hype or Hope” technology. One tool serves as a barometer for this question: Gartner’s Hype Curve. Available to the market since 2005, this curve compares the volume of use of a technology with the volume of communication around it. Divided into 5 stages, ranging from the onset of innovation (1) to the plateau of productivity (5), via the peak of disillusionment (2), the trough of disillusionment (3) and the slope of enlightenment (4), it is precisely the second stage (the peak of disillusionment) that reveals the level of media hype surrounding a technology: I’m not telling you anything new, since December 2022 the hype surrounding AI Generative technologies, and especially ChatGPT, has reached unprecedented heights.
You have to go back to Second Life (2003) to find a technological revolution that has generated so many divergent reactions, from so many different voices on multiple platforms, accessible to the greatest number of people. At the time, the mass media treated the subject with urgency and emphasis, alerting citizens to the duality and risks of a virtual world. Brands did not want to sit on the sidelines and invested heavily. The results were extremely mixed, and today what was heralded as a revolution is still an activity limited to a certain number of insiders (even if the growing Metavers and other hybrid applications – Zwift, for example – have demonstrated some genuinely interesting applicability).
It’s also important to remember that the 3rd phase of the ‘Hype Curve’ (the trough of disillusionment) is always marked by the onset of negative press coverage. This is the moment we are currently experiencing: proven hallucinations (an obviously false answer presented as a certain fact), Guru regret, 6-month moratorium, alarming experiences leading to tragic situations (death) are all proof of this.
The 3U rule (Useful – Usable – Used) is without doubt the most pragmatic approach to technology applied to professional and personal activities. It enables us to distinguish between brilliant inventions and useful innovations, and to judge the degree of acceptance and ease of integration.
In the case of Generative A.I., this rule is particularly relevant because although A.I. has been used in marketing and communication for several years, its active daily and open application was limited to a small number of people.
In practical terms, open A.I. finally offers everyone the opportunity to move beyond theory and come face to face with reality. Described as having no limits other than our imagination, I remembered a pleasant conversation a few years ago with an imminent member of the market (he will recognise himself) who told me that in terms of cycling and music, it was obvious: “you’re a Merckxist-Lenonnist! Admittedly, the somewhat obscure analogy with communist leaders (Marx and Lenin, not to name them) made me smile at the time, but it was totally hypothetical. So, as a practical exercise, was open A.I. finally the tool that was going to help me create the ‘Merckxiste-Lenonniste’ manifesto and iconography?
13 seconds after suggesting to Chat GPT3 that I write the manifesto of a citizen’s movement inspired by John Lennon and Eddy Merckx – the response was convincing, and here’s an extract:
“We, citizens of the world, come together to express our desire for peace, love and harmony. Inspired by the messages of John Lennon and the determination of Eddy Merckx, we believe in the power of the citizen movement to create positive change in our society.
Inspired by John Lennon and his famous song “Imagine”, we dream of a world where peace and harmony are the norm, where borders and barriers are broken down, where diversity is celebrated and where love is at the centre of everything we do. Inspired by Eddy Merckx, we believe in the power of perseverance, determination and hard work to achieve our goals. We are ready to take on the challenges that lie ahead to realise our vision of a better world.”
All that remained was to develop an iconography worthy of this manifesto, and Dalle.e and Midjourney were asked to contribute. This was the most interesting part of the exercise… and also the most confronting. The results certainly didn’t meet my expectations, but I could never have developed them without knowing anything about graphics software.
Was the 3 U’s rule validated? Not completely! Useful: no, in this case it’s futile but fun. Usable: yes, undoubtedly, but it undeniably requires practice – my ‘prompts’ are not yet producing the desired effects and can therefore really be optimised. Used: we can’t predict the level of real acceptance of the tool, but simply commenting on articles and other ‘guru’ releases on the subject won’t give us any insight into its potential or limitations.
In conclusion. The invented answers and hallucinations are real. And ethical questions legitimately abound, but they will only be truly integrated by most people after personal experience. As a result, the only way to avoid disillusionment tomorrow is to face up to reality today. More than ever, testing, testing and more testing to learn, understand and judge is essential!