We recently sat down to chat with Hassan Ragab, an Egyptian multidisciplinary artist and designer based in California who self-admittedly doesn’t know how to describe himself. This is likely due to the sheer breadth of his work which has spanned architecture, furniture design, computational design, and consultancy among many other things.

Since 2022, Ragab has attracted much attention for his conceptual architecture work which he has been posting on his Instagram to a captivated audience. Exploring questions of identity, responsibility and potential, his creations challenge our imaginations and rip through ingrained constraints.

We were curious to learn how generative AI has challenged his own perceptions and what he thinks the future holds. As the zoom camera flickers into focus, we find a bright, engaging conversationalist with a probing mind and singular outlook.

‍

‍

You’ve been involved in design for many years now but when did you first become interested in using AI in your process?

Around the summer of 2022, right after the big models were announced like Midjourney and Dall-E, I guess they were the first ‘big’ models. I was kind of interested to try them out.

‍

Cool. And you work primarily with Midjourney right? Do you use any other tools?

Yeah. Well, Midjourney and Stable Diffusion. But actually, I use a new media process. So it's not only AI. It's overlays with several AI models that are mixed up together. 3D models or images that I make with Photoshop that I use as the base and then I get started with AI. Or the other way around. But for AI generators, Midjourney and Stable Diffusion are my favorite tools.

‍

Nice. Can you talk to us a bit about your creative process? Where do you start off when you have a new idea, or when does AI come into it?

I keep being asked this question about the process but I don’t know... I don't want to call it a ‘creative process’ because it's actually something of a mess. Sometimes there’ll be an idea, and I'll try to pursue it, and most often this idea will kind of transform midway because several inputs get in the way, and I'm not necessarily particular about the outputs that I'm gonna get. I don't really care about a specific visual, I'm more concerned with a feeling. And I don’t know, I think this has always been the case for my work whether I use AI or not. 

But right now I'm really interested in the potential of these tools because there's a lot to discover. And also there's a lot of challenges and a lot of interesting questions that I'd like to answer for myself and maybe for others, like bias and discovering new visual aesthetics and how these tools will affect the future of… I don’t like the word ‘creativity’ but let’s say the future of our visual design, or visual vocabulary. And how this relates to the world we live in right now.

‍

Yeah, well I guess it’s quite liberating when you’re not using it to get a specific output so it’s a more fluid process. And you’re using it as a springboard to explore your own ideas and as you say, to explore feelings.

Yeah. I think it stems from my belief that I don't think individuals are special by themselves so they don't own their own ideas. I think we live in the world and get inspired by many things and reflect on them in our own way. And to have a unilateral view or to have a fixed mindset about a certain thing that you want to create can sometimes eliminate a lot of potentials that can come your way. So yeah it’s liberating, but also it's enriching because basically now you have a lot of opportunity coming your way from a very, very powerful tool that can really enhance your understanding about yourself.

‍

‍

For sure. And I guess because it was trained off of so many people it’s almost like having access to the general pool of humanity. And if you put in a basic prompt, whatever it spits out is basically like the consensus around that topic. Which is incredibly interesting to see visually, you know, what people generally think about the world. It’s kind of like averages right? It spits out the average of what it’s been trained on.

The way I see it is that when I work with AI, it's like I'm working with a reference about the entire humanity. It's like I'm collaborating with almost everybody. Or, like, the documentation of human history. And that's something that I could have never achieved on my own, for sure.

‍

You captioned a photo on Instagram recently that really got me thinking. It was about bias and how AI inadvertently portrays orientalism a lot because of that inherent bias. And I was wondering, I mean, cultural identity is so nuanced, do you ever think it will be possible for AI to accurately portray cultural identity or the nuance of being human in general? 

I think that was kind of my initial question when I started working with AI, but now I'm kind of changing my mind around it because… what is cultural identity? Like you say, there's a lot of nuance. We always look at identity as something that is in the past. But actually identity is dynamic. Identity is who you are, not who you were. I'm Egyptian and I have Ancient Egyptian ancestry and I have Christian ancestry, but I’m all and none of all of these things.

So in order to push our tools in a certain direction, we need to understand the concepts ourselves. And there’s a lot of debate about identity colonialism and orientalism, and these philosophical topics are so detailed you can only observe them, but you can't really figure them out precisely. Maybe there's not even a point in figuring them out precisely. In my work, I'm not really interested in obtaining a certain identity, but rather observing it. Rather, observing the dynamics and how things like bias and things like politics and things like economics affect the way that we use things. Yeah, I'm not really interested in jumping into conclusions.

On the other hand, I understand the value of these principles. But it’s also more interesting to observe how all these factors kind of play together. So yeah, maybe that’s a long answer to the question! But, the bottom line is, I think it's more important to try to use the tools to observe and to document all these questions rather than answer them. To use them to ask questions, not to answer them.

‍

For sure. And when I read your caption I thought, “Wow AI could be a really great tool to reflect our own bias back to us”. Because if it’s been trained on a kind of mass consensus then it’s really interesting to see what it thinks being Egyptian is or being Irish is. It could be a great tool for self-reflection.

Yeah! Also, there are truths in biases. We live in a very fluffy soft world where no one wants to offend, but actually there are biases that are correct about Arabs or about Egyptians. Like, there's lots of misconceptions, of course, but if you really thought about them objectively there is some truth. And, yeah, to embrace bias is another way to understand ourselves instead of portraying this image of like, yeah, everybody's perfect, everybody's great. No, we are biased. We are biased and bias is important for our evolution. Bias is what got us through the last few  millions of years. And bias is important for the training as well. If the models are not biased, they won't be able to produce anything, basically.

‍

‍

‍I mean, it's literally how our brains work. Our brains are always trying to take shortcuts. And just come to the fastest conclusion that it can about information that we already have.

That’s how we survive, yeah!

‍

Since you started using AI in your work, how do you feel it's changed the way that you think about design?‍

Pretty radically. So, I trained as an architect, worked as an architect, worked as a designer for several years. And there's always something that is off with how we do things.There's a lot of good things that come from design and engineering, but there's also a lot of problems. And these problems extend beyond the design itself, they literally affect the way that everyone lives. They could extend to wealth distribution or extend to how the world is being ruled. Everything is connected.

And, as a designer or as, I don't know, a visual artist - I don't know what to call myself! But I’m just really enjoying being in a place where I try to observe this in a very objective way. So I'm really careful now when I'm designing anything because there's a lot of factors that go beyond the design. Like, what’s the function of this? Because while there are tangible functions, there are also several intangible functions that come with design. Yeah, you can build a good design with whatever materials you want but these materials and this design will affect the world.

And now we're living in chaos because there are millions of designers that are trying to create without looking at the bigger picture. So, yeah, I think AI is intriguing, challenging, and its driving my current interest at the intersection of technology, design and culture. I don't want to say it's helping me, because sometimes is also stifling me, but it's changing my mind entirely about how I view the process of creating and everything that comes with that like creativity and usability, and all sorts.

‍

Yeah, it's gonna be so interesting to see how things develop and what the impact will be. Another thing I wanted to ask you is about prompts. What's your process for developing prompts, and what advice would you give to someone who is starting out and wants to develop their prompt process?

Yeah, I don't think prompts are important. So I've been doing this for about two years now, and every time that I achieve a process for prompt drafting or prompt crafting, whatever you want to call it, it gets changed entirely once a new model is released or once there’s a new tool. So you can invest a lot of time trying to create the perfect formula to make a prompt, but this won't help you. That's my personal belief. And even in my workshop, I think the process is all about trying to understand the tool that you're using. Even if you're using Midjourney, there are, like, six or seven different models. Each one has a different way to get the prompt. And also every one of them gives different outputs. So it's very dynamic.

To me, to be honest with you, I think any advice that I would give or any advice that anybody will get will be irrelevant in a matter of three to four months.

‍

‍

‍For sure. Getting to know the tools is probably the most important part. And speaking of tools, I’d love to get your take on the whole ‘AI is going to steal our jobs’ debate. I know you view it as just a tool but do you think it will evolve beyond that?

Oh, yeah, it will always be a tool. I think the question about AI replacing people… it will only replace people if people are tools themselves. And that's what is happening in most of our industries. Like, if I'm talking about architects or, visual designers or, like, graphic designers or whatever, you can see there are generic rules that everybody is really literally following to the letter. And you have the same process in Hollywood. You know, they have a manual for how to write a script in Hollywood. And everybody follows the same formula.

There's one funny thing that I heard over a podcast a couple of days ago. Now they are designing Netflix series, the term was, like, a second screen, meaning that you have to design or direct the show, assuming that the user will be on their phone and that will give you a second screen. So that means that you have a limitation for the number of frames per second or, like, you know, your scene can be very slow, can be very intense. You seem all these factors are factors that you feed to a machine, not a human being!

And in that case, AI will easily replace all of those people who do not think for themselves, who are only following capital or only following directions from owners. So in that case, 100% AI will be able to do all of this.

Another thing is… it seems like nobody is really focused on what's happening right now. And everybody is very excited about the future. But we have problems that we need to look at right now. If we're not solving them, if we don't think about them thoroughly, the future will be catastrophic.

We live at a very interesting point in time because there are so many factors that are really connected but nobody wants to look at them. Everyone is thinking, will AI take my job? You might not even have a planet!

So yeah,  basically  my two takes are: people are already tools right now. And the world is bigger than somebody’s job.

‍

Exactly. But also, the optimistic side of that is like, maybe that would be the very thing then that will force more people to become truly ‘creative’ as a reaction to AI replacing people as tools. That there will be more auteurs in cinema, for example, or there will be more people breaking the mold and it will push people to create human work that AI can't create.

That's also, of course, something that we need to look at. Like, when they invented photography, everybody felt like this is the death of art. But actually, I was talking to an artist friend the other day and he told me, like, the invention of photography liberated visual arts.

‍

‍

‍Is there a tool you’d like to see developed?

Well, there are a couple of things. So one of the things that I want to do myself is I want to build a model that is customized on Egyptian visual aesthetics. So like Egyptian Architecture or maybe Arabic. I don't know. But I mean, to get  a model as big as Stable Diffusion, but for the Arab world. I think it would be a good way to do visual research to understand questions around identity.

And also, if we're thinking of a central AI model, I’d like to see something that would try to push people towards being more empathetic. Or to help people have a better life. So I was working with a tech company in San Diego, and I was talking to the CEO, he was a doctor, and he was saying that sometimes technology can help us achieve our goals. So, say you're walking in the street, you could filter which ads that you want to see. So instead of seeing a big McDonald's sign, now you can see that you need to have a salad or you need to drink water. 

Or maybe it could show you… instead of seeing billboard ads to buy a new home and or a new car, maybe to remind you about what's happening in the world and try to be empathetic towards other people.

Or, this was another friend of mine's ideas. He was talking about how AI could be your own personal assistant to become a nicer person. So, basically, it follows your interactions every day and then it reminds you, like, hey, you haven’t been so nice today or you didn't call your Mom. So it's like, your own consciousness that's actually kind of speaking to you. 

I hope these things will be implemented, I really hope AI can help us out. But we need to think about how tools and people are being weaponized against each other. Those are the fundamentals that we need to think about before we build our tools. It’s like nuclear energy. Yeah, it's a very powerful tool, but it kills…

‍

This is the thing. I mean, we have to hope that it would be used for good, but given human history, it doesn't look very likely…

Yeah, yeah, I mean, I'm always a pessimist, but I think… there's no virtue in being a pessimist. And I think we always have to hold on to hope for a better day. So, yeah, I mean, we'll see. At least we'll experience something that's interesting.

‍

Yeah well all we can do is hope for a better future.

We’ll try. We need to try.

‍