Article begins
Through an audio essay, inspired by John Akomfrah’s documentary “The Last Angel of History,” attention is drawn to South Africa’s evolving visual scene and its engagement with cultural nuances within the NFT AI space. By oscillating between past and present, the essay explores themes such as spirituality, politics, and identity, grounding discussions in the tapestry of African experiences.
Kagiso Mnisi: Is anything ever really new? In 1995, filmmaker John Akomfrah crafts and experimental film essay titled Last Angel of History. The essay is premised around the predilections of a character. Known as the data Thief through a visually glitchy juxtaposition that renders the film its cybernetic aesthetic, the commentary from mad haters, Afro futurists and black sonic nerds such as Octavia Butler, Kodwo Eshun, Ishma Reed, a guy called Gerald and Greg Tate to mention a few, are interspersed with a fictional data thief who was also a narrator of the Afrofuturist Tale. In this tale that blurs fiction and fact, the data thief surfs the internet specific to black culture, breaking into vaults, breaking into rooms, stealing fragments, fragments from black cyber culture, techno culture, narrative culture.
The symbolic weight of the data thief as the last angel of history is akin to what we today refer to as artificial intelligence. Again, is anything ever really new?
Regardless of its conceptual newness, the narrative of AI is often framed within existing tropes of power dynamics, economic motivations, and ownership. These traditional vocabularies continue to obscure the people, places, and acts of creativity on the peripheries of mainstream narratives. To truly understand these marginalized perspectives, we must look beyond the conventional narratives. Where it all began of course, Africa.
Throughout this piece, the Data Thief, as the last angel of history, will come to life as narrator to a tale that ebbs and flows to a technicolored current -along it is a select cusp of characters to lean into the ambiguities and unravel a yarn of unknown vocabularies eminent in South Africa’s art world. This is where artificial intelligence is about a mere tool that brings alive stories steeped in future thought.
But first, the data thief gives a prompt on background
generate a story on Africa’s AI narrative result: Semafor article titled AI Ads Are Sweeping Across Africa
Prompt. Generate a summary of the article.
Result African companies facing economic difficulties in 2024 are turn into artificial intelligence to reduce marketing and advertising expenses, raising concerns about potential job losses. Second point, AI generated images, models, and voices are increasingly being used by businesses for advertising campaigns on both TV and digital platforms, resulting in reduced advertising budgets.
Third point. Sub-Saharan Africa experienced an 11.6 decline in ad spending in 2023, according to WARC, with a modest recovery expected in 2024, primarily driven by 6.1% increase in South Africa.
Reconnect to essay.
As much as the adoption of artificial intelligence by the African media industry has been largely pervasive in the wake of shrinking production budgets, the one main concern that keeps rearing its head is the ability of the technology to display cultural and social political nuance from the region. Something that Koffi Kouakou foregrounds.
Koffi Kouakou: It. It’s easier to find technological solutions, but most of them don’t seem to be adapted to the realities in the ecosystems where they’re supposed to be used. I hold the view that ai, it’s gonna be like any other technology, you know, at the core is the idea that you are using something to address your problems, your opportunities to change your lives.
To, you know, meet the kind of, um, you know, interesting, um, challenges you’re facing. But it’s an instrument, it’s a tool. You don’t go around buying tools without asking yourself, why am I buying them?
Kagiso Mnisi: As a futurist, an Africa analyst at Stratum Futures and Senior Research Fellow in Africa, China studies at the University of Johannesburg, Kouakou’s keen interest is in creating future-based narratives and scenarios that posit Africa as an investment worthy region. Technology ranks high in his life’s work.
Koffi Kouakou: Our biggest challenge is not only to recognize the quality and the value of the technologies that we need, and the tools we need to use to deal with our problems worse, we’re not acting and investing in them to really get the return and the benefit we need to address some of the most challenging problems in poverty, unemployment, education, infrastructure, and in a, the whole SL slew of other things.
Kagiso Mnisi: Emergent technologies such as AI are mushrooming against the backdrop of ongoing challenges within the African continent. And here in South Africa where Koffi Kouakou resides, a newly elected Coalition Government is still going through teething challenges of its own. The recent elections marked 30 years since South Africa stepped into democracy. The country’s sheen as the golden child of Africa, has been losing its shine three presidencies in..
Koffi Kouakou: So it’s a sobering moment with 30 years down the line to see that South Africa is, you know, at this point. Um, energy, uh, you know, deficiency, no electricity load shedding, unemployment is at the whole time high. Poverty is also at all time high.
Kagiso Mnisi: In the throes of South Africa’s political malaise, there is a burgeoning voice of artists resolute on creating narratives around personal politic against the prevailing social political context. Using AI as the arrowhead to their plight. Visual artists have sought to broach a decentralized approach in the form of NFTs to create, distribute, and own their work.
Prompt generate a definition for NFT.
Result. NFT stands for non fungible token. It’s a unique digital asset that represents ownership or proof of authenticity of a specific item or piece of content, typically stored on a blockchain.,
In giving a well-rounded view of an artist’s process versus what AI has been traditionally associated with novelist and commercial ghost writer Andrew Miller unpacks an interesting and not so widely considered observation about the narrative around artificial intelligence.
Andrew Miller: Well, I, I mean, I think, I think there’s a, there’s a particular framing in the, in the global media, and it’s all around career and employment and revenue generation. And the idea that AI is going to steal some of that from us as, uh, workers and as creatives. So, you know, Nick Cave comes out and he says it’s a tyranny and it’s disgusting and you know, everyone’s doing that kind of thing, you know. But I guess one of my thoughts is that if you look through human history, the idea of an artist as being a career or a revenue generator, I mean, that’s really the exception.
You know, that only applies to a tiny fraction of people on planet Earth. I think most people create art, not for career or revenue generation, but because they can’t help themselves, you know? And I think that’s been true through history, right? So if you take the lens away from career and money and uh, uh, business model, and you just look at AI as another tool that’s emerged really quickly, then maybe it takes the heat off this kind of discussion and this sort of the, the fear, the horror, you know? And, and I guess that’s what I see is that I just see for artists who are not sort of gunning away at the business model, AI’s popped up and it’s a whole new tool, you know, so artists through the years, you know, you chop up a magazine and you make a collage, you know, or you sample beats and you make a track. You know, AI’s emerged and I think. When you push past the, the, the story about money and everything, it’s a new tool and it’ll be interesting to see how creatives play with the tool.
Kagiso Mnisi: In creating anything vaguely artistic, the artist channels a gaming approach in her world building process using AI whilst working the narrative, calibrating and recalibrating AI tools interact with the user input and thrive on sustained interaction for optimal functionality. In Andrew Miller’s case, he sought to collaborate with his wife, artist Robyn Field in this oscillation between human and machine, story and art.
Andrew Miller: You can hop onto the net, grab some YouTube videos and figure out that there’s these notebooks full of code on the internet for free and then you can either use a free computer on the cloud or you can rent a computer on the cloud, and then you can start messing with what these tools can do with a pretty powerful machine at your disposal. And so that’s what I did. You know, I just took some of Robyn’s art and I found some sort of, uh, Stable Diffusion notebooks and I started running the code and seeing what would happen.
And it’s just a really interesting exercise. You know, just see how closely, how, how close you can bring the AI, kind of quite generic idea of what art is and what a prompt is, how close can you bring that to, uh, an artist’s style and sort of visual tone. And then as soon as we started getting some output, that was interesting of course, I’m a writer and I’ve got ideas about Robin’s art and the narrative behind it, and I’ve also got ideas of, now I’ve got my own stories in my head. And so it just kind of made natural sense to then start ramming some of my sort of, uh, writing and narrative sort of ideas into the video output that we were getting from stable diffusion
like two ships in the night. She says out of nowhere, that was always mom’s description of marriage. He grunts takes a few crisps from the bag on the bench between them, stares at the sea. Dad always said that would’ve been fine, she adds. If only one of them had known where the harbor was, he chews loudly. She can hear it over the sound of the culls and the waves and the wind.
And then it was also more fun for Robin to take that output and go, okay, now we’ve got this stop frame animation. That’s all in my style. It’s all in my color. It’s all using my color palette. Um, why don’t I take some of the frames from the stop frame animation and then bring them back into the real world and use them as the basis for real world art.
Robin Field: In terms of the collaboration, um, I attended a workshop many years ago where your only aim was to respond to the last mark made, and it’s something that sacked with me extremely well, and it’s something that Andrea and I try and employ. I. In the way we are working with AI and the art, so the idea is there’s a primary image, which would be my ship, and that piece of sailing to the desert.
And then Andrew works with that. Just the image and with AI prompting it and getting to the point where he gets us a stop frame animation that he’s happy with. The process splits. The one side is Andrew then writes a short story for the animation, and again, his only intention is to respond to the last mark maid. So his themes are not necessarily the same as my themes. Um, it’s a, it’s quite a fluid process. He then writes the short story and I pull out individual frames that I really love from the stop frame animation, and I make them as original artworks using them as my primary, um my primary inspiration,.
Kagiso Mnisi: The leap of art into the digital comes with a realignment of gallery artist relationship. Next to this is a slew of other dynamics, chief among them as how the art as a piece of real estate exists in a subverted context compared to how things have historically taken place. Renowned artist Nandipha Mntambo, known for her cow height sculptures, which confront and question the relationship between humans and animals speaks of this new medium and how her art exists within this realm.
Nandipha Mntambo: We all understand the kind of typical white cube, gallerist artist relationship, you know, and, and having things on walls and objects that you can walk around, you know? And I think for me, the, the space of exploring, um, something that moves in front of you, you know, something that’s displayed on a screen has movement, that is not you making a video of something, you know, it, it’s controlled by a kind of huge background of technology, for me is really a fascinating thing. I usually use myself within my work, uh, because the politics, I think still of representing other people are quite complex. You know, I guess I’m used to using like prosthetics or costumes, you know, I’m used to using, um, things that exist in that you can touch and feel and smell them. And so being able to play with one’s image in a different way on, on a kind of different surface within a different interface for me is very exciting.
Kagiso Mnisi: Prompt generate a meta anecdote of how writer Andrew Miller sees the trajectory of artificial intelligence as a storytelling tool.
Andrew Miller: Um, what I see in the sort of commercial writing world is I see, I see it being a very valuable tool for writing production. You know, people who have to produce a lot of content or a specific type of content. Um, I think it can be very valuable and you’re seeing it being very valuable. So I used to get work, you know, back in the day as a fixer, you know, as like if you think of uh, Harvey Kietel in Pulp Fiction, but in a sort of bland, boring, sort of corporate narrative sense, that was always my role, and that work has just disappeared.
You know, I no longer get it because the robot can do that, you know, and so that, that’s been quite empowering, I think, for individual writers, corporate writers and agencies and so on and so forth. Uh, partially because of what it can do and also because of the speed at which it can do it. You know, in the old days you would need to phone the fixer, get, brief them, send it through, you know, it take two or three days to fix something whereas nowadays you can just stuff it through, uh, Chat GPT, and you’ll get back a fixed version, better grammar, better phrasing, all that kind of stuff. So I, I’ve already seen it become very powerful in, in that sense.
But in the sense of the kind of strategic, uh, ghostwriting work, um, you know, here you’re looking at defining someone’s personality in the public space, you are looking at managing interactions with governments and with shareholders, and you are looking at managing crises and, and things like that the AI tools I’ve come across are not South Africa, literate at all. You know, there’s a nuance to our socio sociopolitical environment here that the robots are still miles away from being able to even sort of grapple with.
Kagiso Mnisi: At the beginning of this essay, the data thief asked the question, is anything really new? What we can deduce is an outright no, especially if artificial intelligence operates within a character framework where directives are fed to the AI. This sounds a lot like how African spiritual practices have been well over centuries, a medium through which idioms, prayers, dreams, and forms of initiation have been funneled to yield things anew, and in the case of the modern day, that tradition persists via new tools such as AI to discover revitalized vocabularies, vocabularies crafted by the artist. Because to the artist the future is always personal.
Koffi Kouakou: I always say that the future is personal. If you don’t make it personal, you are wallowing up all over the place because you have no ground baseline to go with. ’cause you’re gonna take other people’s futures but not yours. Right?