"I’m a prompt man"
Relationships with artificial intelligences - Media Ecology Newsletter #32
Water scarcity is a challenge for ecology, economy, culture. It is not the subject of this issue, but it is such an important matter that we cannot be silent about it. We need a new ability to design in complexity to develop the right policy about water, an ability to think of it in a completely new way. If humans need to think anew about water, and many other subjects, also a new media ecology is needed. If you are interested, take a look at “Scritti sull’acqua”, with some notes about water (in Italian).
--- “I’m a prompt man” ---
I interviewed many years ago the head of research at one of the companies that made Internet history in the 1990s. We were talking about how he organized the research team. He told me that it all depended on how you hire researchers. And he told me that he had a method. At that point he sort of interrupted and asked me.
"Do you know the legend of Camelot?"
"Of course," I replied.
"Which character did you want to be?"
"Parsifal."
Hearing that, he widened his eyes and told me that I challenged his method.
"Why?"
"For years I've been asking the same question to everyone who wanted to come and work in research. You know Camelot, what character did you want to be? And everyone always answered me one of these three characters: Arthur, Lancelot, Merlin. And I always kept in the research only those who answered Merlin. And I've always had good results. But why did you say Parsifal?"
"Because he set out in search of the holy Grail. And the first night he happened to be in the castle of the Fisher King. During dinner, he saw two people in white tunics carrying a plank on which there was a cup. He had an intuition, but he felt embarrassed to ask if that cup was the Grail. And he did not ask. And not having asked the question he had to search for decades more. I feel the same urge to ask all possible questions."
And that is the job of the researcher. To ask questions, first and foremost. But this becomes the strategic job in the age of generative artificial intelligence: what matters in this context is to make the right prompt.
In this issue we are exploring the importance of “questions” in the new world of artificial intelligence. In order not to be trapped in a self-referential world, we need to keep our eyes open to reality: we need to be open to seeing what is possible even if the paradigm in which we are immersed does not let us see it. The key moment is when we ask the questions. Do they come from our experience or are they derived from the tool in which we think the answers can be found?
I hope you will forgive me for using the word “man” in this issue titles: that’s because I’m kidding with the title of an old song: “I’m a soul man”. And adding soul to the question of artificial intelligence is exactly the point of this issue.
--- Well, yes, “I’m a soul man” ---
There are 14 million songs that have been generated by artificial intelligence users offered by Boomy and uploaded to streaming platforms. These songs compete with those written by human artists for traffic and attention on those platforms. How would listeners' experience change if they knew they were automatically generated? According to the Financial Times, Spotify removed tens of thousands of Boomy songs because their ratings were inflated by an audience that was itself made up of artificial intelligence and bots. The ultimate in self-referentiality: music made by machines and listened to by machines. (ArsTechnica, La Svolta).
Perhaps soul needs humans to make sense. But certainly, automatically generated music will not stop. Rather, it will flood the platforms, form the background of videos, become ambient, and so on.
In this context, the artistic problem shifts from composition to prompt.
--- How to think a prompt that works ---
The magic of finding a good prompt is like the magic of finding a new dimension of what’s possible. The prompt becomes the real creative act in the relationship with artificial intelligence.
If one wants a rational approach to it, which is not the creative approach, but at least serves to orderly address the issue, one can find a sort of knowledge base to define the structures of questions making that comes from the experience of what works with large language models. With this approach, prompts are specific and directly aimed to the goal: “summarize” can help with large texts that one wants to become short; and “topic” can tag the texts; define the output structure can also help with generating a reliable text. Microsoft offers some pages with instructions about this kind of approach, such as: “Introduction to prompt engineering” and “The art of the prompt: How to get the best out of generative AI”.
I also did a different research prompting the AI with a simple question in different languages: “Who was the most important poet of all times?”. The beauty of the question was the ability to test the link between the answer and the language: it turned out that ChatGPT was very much centered on the Western and specifically American culture, while Bing’s chat was more open to different cultures. It is a research that could be explored further.
Please take a look at Reimagine Europa. A Media Ecology Research Network is being build in Bruxelles and it grows every day. I will be informing on that more in the next issues. Reimagine Europa.
In previous Media Ecology: April 21th – The truth about fake April 7th – CheatGPT March 31th – Artificial intelligencija March 16th – The method is the message February 9th – Epistemology of AI
Posts in Oecd’s Forum:
March 17th – GPT4: the hallucinations continue
February 21h – A Tsunami of Illusions: An artificially intelligent trompe l’oeil
February 8th – - Knowledge ecology of the new search engines
Podcasts in Italian, by me
L’altra metà del verso. Rai Radio 3 Media Ecology. Intesa Sanpaolo on air Eppur s’innova. Luiss University Press
Ecology of screens
On the occasion of the International Conference Vivre par(mi) les écrans: entre passé et avenir, which was held in Lyon at the end of May, the newsletter of the International Research Group Vivre par(mi) les écrans and the Media Ecology newsletter agreed to signal, each to its recipients, the importance to them of the other's content, inviting them to subscribe to receive it and disseminate it among their contacts. So please visit Vivre par(mi) les écrans and subscribe to the newsletter. This choice of collaboration stems from the common project of promoting, developing and sharing highly qualified knowledge aimed at creating tools for guidance, critique and intervention in the field of media ecology and our current and future living between(mite) screens, as well as fostering the social dissemination of the aforementioned knowledge and tools.