10011.coServicesProjectsGet Started

Aarati Akkapeddi

Interview with Aarati Akkapeddi

Riley Gunderson
Production / Direction

Aarati Akkapeddi is an artist, programmer, and educator living and working in New York City. Their work discusses the politics and poetics of data sets through the use of machine learning technology and visual archives. They are specifically interested in how different ways of organizing, preserving, and working with data affects how we perceive ourselves and others.

From the series, "Ancestral Apparitions," September 2020, Code, Lumen Print

R: When did you first start working in this way?

A: Probably in college. I was interested in working with data and coding, but I don’t think I got to my specific interest in archives and specific images until pretty recently, probably in 2017 or 2018.

I have always been really interested in photographs. I guess I just hadn’t connected my practice as an artist and my interest in family photographs and preserving them until around that time. I was in grad school from 2017 to 2019. I already had been thinking about identity in relation to data and how we take images now, but then I started getting interested in documentation. I had always been an unofficial family archivist. I would always save photographs from my grandma’s house and scan them so we’d have digital copies. I guess something clicked there and I finally started working with those images.

R: Have you seen your work evolve in any specific ways over time?

A: When I think of earlier work that I made, I think I was afraid to put myself into it in the way that I do now. I think that it is hard and easy to create work that is personal. It is easy because it comes naturally to speak from your own point of view. But I think it is difficult because it is more vulnerable and you are putting yourself out there more. For me, that has been more fulfilling and something that I want to continue for a long time.

From the series, "Ancestral Apparitions," September 2020, Code, Lumen Print

R: Your work discusses identity a lot, how does your own identity influence or impact your work?

A: I was born and raised here in the States but my family is from India, both of my parents came over in the late eighties. I think that also has to do with why I’ve been, even as a kid, really interested in photographs. They felt like connection points to my family history, especially when India is literally so far away. It felt very grounding to me to have some kind of visual marker of stories I heard or people I heard about. I think in working with them I am also interested in their aesthetics and thinking about how we represent ourselves, families, and family structures and how that reflects our changing values.

R: Has your relationship with your family photos changed since you started working with them?

A: Yeah, I think so. For my graduate thesis I worked with my family photographs but I also worked with a photo archive in India that has been collecting studio photography. A lot of those photos are family photos, just of other families, but they were taken in the same time and place that a lot of my family photographs were taken. As part of that research I interviewed a lot of archivists here. That made me get really interested in the ephemerality of our images now and thinking about digital preservation. To be honest, it kind of scared me a little to think about our images now and how easily things can disappear, and even how now we rely on platforms like Google Drive and Instagram to hold onto our memories.

R: Can you talk about your process of combining digital and analog means of working?

A: It's super fun. It is really interesting to me to return something to its medium, especially with photographs. Recently, I worked with machine learning and family photos to produce digital images, then I used those digital images to create negatives to expose photo paper with. I was interested in doing that last step because it brought the project back to the medium of the photograph. I also used expired photo paper which gave it the same quality that the original photos had, the photo paper I got was from the same era, the seventies and eighties. I am also really interested in printmaking and alt photography techniques like cyanotypes and anthotypes.

From the series, "After Image Risograph Prints," Gold risograph prints of computationally averaged similar photos from the Studies in Tamil Studio Archives and Society

R: Similarly, you work a lot in a variety of mediums, how does your creative process help you navigate which mediums to use? Do you work by trial and error or do you like to plan things out in advance?

A: I think there is definitely a lot of trial and error. I try to think about the medium in relation to the work itself, kind of like what I was saying about the photographs. I think there is a back and forth between trial and error and research. I research the threads between what I am trying to say and then how the medium affords that or doesn’t.

R: Your work deals a lot with archives, can you talk about your process with archival material? If you are sourcing from external archives, how do you work to make the content your own?

A: It has been extra hard with COVID. New York City has a lot of great resources in terms of archival material, even just for inspiration. I really think it helps to be physically there, but obviously I can’t do that these days. Actually, in working with that archive in India, I did the majority of that work long distance. I think reaching out and talking to people really helps. It can be really inspiring and give you a better context for what the archive is about and how it relates to what is important to you. Also, archivists love talking about what they do. They are really smart people.

With The Studies in Tamil Studio Archives and Society, the archive I collaborated with for my thesis, I reached out to them to get permission to use their photos. I think that is also really important. Although there are a lot of publicly available archives, give credit where credit is due. I do think that is particularly complicated with machine learning work because you gather all this data, usually thousands of images, and then you feed it into an algorithm and it spits something out. It is easy to obscure the origin of the training data so giving credit is really important.

R: I find it interesting how personal your work is even though you employ technologies and subject matter outside of yourself, can you talk more about that?

A: I think working with archival material is really interesting in terms of finding throughlines with yourself. Looking back--especially in terms of my relationship to family images, or our relationship to photography historically--has made me rethink my personal relationship to images now. I haven’t necessarily made work about my current relationship to images but it has influenced me in a lot of ways. I just deleted my Instagram and backed up all my photos in like three different places. I think a lot about digital preservation. I think looking at archival materials and letting creative research take you on these different threads of connection can really help you learn about yourself and what is important to you.

From the series, "Recollections in Haldi & Light," September 2020, Code, Turmeric Anthotype, Lumen Print

R: You seem to work with a lot of complex technologies and programs, how did you learn about them and how to incorporate them into your practice?

A: I guess it all started with me learning how to code in college and getting interested with incorporating that into my work. With machine learning, it was definitely a slow process. It was hard in the beginning because I had to teach myself how to work with it, and also how to gather so much data to train it. Since I have started working with it, I have found that there are actually a lot of tools out there for artists to work with. There are two companies I can think of, Runway ML and PlatformIO, that both have software you can use without needing to know how to code. I feel like it is becoming more accessible, it is just taking time. I think the hardest part, and the most interesting part, is just finding the data that you want to use because you usually have to find a lot of it.

R: I was looking through your website and you discussing these technologies and artificial intelligence in a few different ways. Sometimes I found it hard to distinguish if you were praising or critiquing artificial intelligence, can you talk more about this?

A: I think I am also a little confused whether I am critiquing or praising artificial intelligence. It is not that I think that the technology itself is bad, I am just very wary in the ways in which it is being used. My stance is that we need to be really aware of what data we are using to train machine learning on. My work is about that data, about the archival material and working with it. What I don’t agree with is the obfuscation of the data that is being used. For example, with facial recognition, it is said that the technology exists in a vacuum and is perfect, but in reality they trained facial recognition with like ninety percent white male faces. I believe in being transparent about that, and not using it to replace social and political infrastructures either.

R: You mentioned a few times how social media has influenced your work and your relationship to the image, can you talk more about that?

A: This is more of a recent revelation and I am thinking about spending more time with it. It was really important to me to have those analog family portraits and we don’t really take photographs like that anymore. The way we choose to remember moments is so different because of social media. I just deleted Instagram and I had to manually download all my images because the download feature didn’t work for me. I spent three days downloading images and backing them up. I don’t think everyone would do that. What if tomorrow Instagram was bought by some other company and you lose all your images? What does that mean in terms of ten years from now and you don’t have those images depicting important moments? I also think the way images look is influenced by how we think of them in terms of volume, too. When you take an analog photograph, you take two or three images. Now, you have like ten thousand photos of your dog at the same angle.

From the series, "After Image Risograph Prints," Gold risograph prints of computationally averaged similar photos from the Studies in Tamil Studio Archives and Society

R: I noticed a lot of silhouettes in your work, are you interested at all in anonymity in regards to identity?

A: I am really interested in silhouettes. Not so much in terms of anonymity, although I can totally see how that comes through, but more so missing pieces. I am thinking about my identity, the fact that I am part of a diaspora, or diasporic identity in general, and finding these connections but also thinking about the connections that are lost. The silhouette felt like a good communicator because you know there is something there yet you can’t quite figure out the details. That is how I feel when I am trying to reconcile my family history and both parts of my cultural identity, being American and Indian. The silhouette has been a really good metaphor for me.

R: Do you have a favorite project that you have worked on?

A: I really liked working on this project, Ancestral Apparitions. I would like to continue working with expired photo paper because it is a really fun and interesting process. There was something that felt really right about starting with photographs from the seventies and then returning it to actual, physical material from the seventies and eighties. Also, working with light in that way. It felt really good to pull light out of the screen and back into the photography process of using light to make an image.

R: What are you working on now?

A: Right now I am working on a project with CHAT, it is a group exhibition curated by Taeyoon Choi at a textile gallery in Hong Kong. It is a project working with machine learning and archive, but this time looking through and reforming textile patterns by training the machine learning model on them. They are textile patterns specifically from India. It is still in the early stages but I am thinking about memory and motif. I am also working on writing more because I feel like it is really helpful to organize thoughts.

From the series, "Recollections in Haldi & Light," September 2020, Code, Turmeric Anthotype, Lumen Print

Aarati Akkapeddi: https://aarati.me

Interview By Riley Gunderson

How does this sound?

We manage all your social media accounts and you focus on your business.

Thank you! Your information has been received!
Oops! Something went wrong while submitting the form.

Have a project idea?

Start here.

email us
schedule a call