Strategies for ethical systems and organisations, and for empowering AI artists: with Eva Jäger

Photo by Boris Camaca

At The New Real, we want to reflect and represent those at the forefront of art and AI – and to work with them to develop actionable strategies and signposts for practitioners.

Here Eva Jäger, Curator of Arts Technologies at Serpentine, looks at ethical and relational business models, and emerging practices of artists.

We wanted to know: What are the viable alternatives to current extractive AI business models, how can we achieve them, and do they work at scale? What are the common issues at the heart of these debates; what does the current and emerging generations of artists want; and should we even think about intelligent art through a business-lens? In short: How do we build ethical, fair and prosperous systems and organisations for art, audiences and artists?

Here is what Eva told us.

– 

We need a multiplicity of alternative ways to experiment with AI to explore the full potential of this emergent intelligence.  

As a society, we have set up many different ways of building, organising around, governing the technology we create––creating a saleable product ready for market is just one way to engage.  However, what this technology is showing us (many important technologies have done this) is that they can evolve to do things we didn’t expect. Think of DeepMind’s AlphaFold––an AI which is able to solve the ‘protein folding problem.’ It was an evolution of the same AI that was known for learning to solve all the Atari video games. The point is that we have to recognise that we are still in the imagination space of AIs development and if we are not too quick to control it, but instead learn what it can do well and evolve it to work to do those things with us, we will be much better off.

Let's look beyond data sovereignty towards a more relational view of how cultural production happens.

I learned from Matt Prewit and also Salome Viljoen’s work about the fundamental relationality of data (personal info, things we make/produce, and data ‘exhaust’). We are in a current moment where ‘data sovereignty’ is dominating conversations––I own my own data and I should have inalienable rights to it. However, their work has helped me understand that the individual is the wrong level from which to bargain for data rights, even for artists. Since data is valuable only in relation to other data or in collections of data (since it’s used mostly to train, analyse, predict it needs a lot of data to make generalisations) ‘my own’ data is not really worth much to others in isolation.  However, if if we start thinking about bargaining from the entangled and relational point of view of groups of data (like collectives, coops, daos, trusts, etc.) forming around specific kinds of data, we can start to see that those kinds of organisational forms might have a lot more power especially if they had regulatory backing that assumed data users couldn’t get data for free. 

New business models and new forms of data exchange are needed to counter our new data medievalism.

An artist/technologist I work with, Mat Dryhurst, explains the above really well by saying that our data relations are becoming mediaeval––everyone is building moats around their data (x, Reddit, etc.).People, organisations and companies realise that they’re sitting on really valuable data resources though that wasn’t the core of their operations to begin with. That goes for public institutions as well. I recently interviewed Theresa Züger who runs the Public Interest AI Lab at Humboldt Uni in Berlin––she explained that a whole new organisational form is going to be necessary for public institutions to develop operations around their data to make them more useful and accessible, almost like service providers. This is going to be a really interesting moment for data governance as it is going to touch every aspect of the public sector.

We need to be open-minded to Web3 solutions, and see them for what they are, as well as what they could be.

We’re in an interesting moment for Web3, where the arguments being made for ‘trustless’ verification systems  have a real utility for data and AI model verification. We also know that crypto is super alienating and clunky. My take is that we need to engage with blockchain affordances openly since there’s a lot of really interesting alternative models for value distribution for cultural production and data exchange. 

We must not be fooled into thinking that technology is changing the world at break-neck speed.

If you read Twitter too much, you would be forgiven for thinking technology is changing everything right now, but that’s not the reality. It takes a long time for technologies like AI and blockchain to seep into communities and be trusted – we still use ballot boxes after all. 

We need to focus on the core values we’re aiming for, not the technology itself.

Technology is just a means to get us somewhere, along the way we may have to adapt our strategy until we get to the point where we’re making good on the core values we’re navigating towards. In the cultural sector we’d be better off not going for the hype of a particular technology But rather using technology to create the kind of change we want to see, demonstrating that technology can be a creative and social medium.

We need to shift our exploration of artistic work from artefacts to an artistic system.

This is true for architects, designers, writers,visual artists, and so on. As we move closer to a world where generative images, audio, and language models can produce evocative content ad infinitum, artists will increasingly identify their ‘artwork’ with their own creative tech system including their own AI models and databases.

We need to think less preciously about the end-piece of art, and think more openly about what it could mean to ‘fork’ artistic output.

If your cultural production is a system, what and who would be part of that system? What might it look like to join your system with another? This shift allows me to think about  the whole creative process and not just the final artefact. It also allows me to imagine interventions or forks that could diverge from me as the only ‘agent’ in my creative process. Holly Herndon’s ‘Holly+’ is a great example of this, it’s her synthetic voice and it’s available for use but it’s subject to authentication by the Holly+ DAO which includes those involved in training and developing the tool. 

We need to break down the idea of the individual artist genius, especially now we have different forms of synthetic intelligence to play alongside and with. 

The idea of the ‘individual artist genius’ is super problematic––it’s not reflective of the collaborative nature of art-making, especially with tech. And it’s not reflective of the way culture spreads through networks, gaining value through all the nodes of participants. AI has the ability to show us collective production and also to be another kind of collaborator Now that we’re living with  artificial intelligence, it’s a good moment  for self inquiry into human intelligence as distinct from individual ‘genius’ 

We need to remember that human creativity is not under threat.

I am open to the idea that the shape of what it means to be an artist is going to change. With openness, rigorous inquiry and experimentation we can have a say in how that unfolds. The modes of creative production might change but our urge to communicate through art-making isn’t going anywhere for us humans.

–

As Curator of Arts Technologies at The Serpentine, Eva Jäger commissions artists working with advanced technologies and collaborate in teams designing novel approaches, workflows and philosophies of emerging tech. During her time at Serpentine she has worked with artists Holly Herndon and Mat Dryhurst, Jenna Sutela, Hito Steyerl, Suzanne Treister, Jakob Kudsk Steensen, Trust, 0rphan Drift, Kite, Keiken, Danielle Brathwaite-Shirley, Libby Heaney, Gabriel Massan and dmstfctn. Eva is also part of the R&D Platform both as Co-I of the Creative AI Lab and also as part of the team (lead by Victoria Ivanova) producing Future Art Ecosystems (FAE), an annual strategic briefing that provides analytical and conceptual tools for the construction of 21st-century cultural infrastructure: the systems that support art and advanced technologies as a whole, and respond to a broader societal agenda.

These strategies were formulated/articulated/conveyed in an interview with Eva and have been edited for clarity.