On the last Friday of each month, we spotlight a CLCF researcher with our profile series (check out our previous profiles here). This month, we’re delighted to introduce Sanaz Mazinani, Assistant Professor of Studio Art in the Department of Arts, Culture, and Media (UTSC). Sanaz Mazinani is an artist, educator, and curator based in Tsí Tkarón:to/Toronto. Working across the disciplines of photography, sculpture, and large-scale multimedia installations, Sanaz creates art projects that invite a rethinking of how we see and experience knowledge. In her work with Artificial Intelligence, Mazinani investigates and critiques its use as a creative image-generating technology, exposing the pitfalls and biases of the technology and how artists need to inform the technology’s future.

M.E.: In a nutshell, who are you as a researcher?

Sanaz: As an Iranian-Canadian a great deal of my research is informed by my positionality with regards to East-West relations. My practice critically examines the representation of conflict in news media and how these skewed representations inform identity across borders. As an artist, I translate and output my research through photography, large-scale multimedia installations, and social sculpture. For example, when my work is about the destabilizing effects of media in portraying conflict and war across cultural divides, I employ artistic tools such as repetition and pattern to transform information and challenge viewers’ perception of what has historically been accepted as normal.

As an activist my work invites audiences to interact and narrate their own experiences. The viewer is an essential element in completing the experience of my projects, which are frequently interactive and immersive in nature. For example, a recent project titled Threshold uses the reflective surface of mosaic mirrors to invite people to see themselves directly in the work and hence implicated in the subject of the videos and sounds that surround them about explosions.

Daphne:  How do you see the role of artists in shaping the use of generative AI?

Sanaz: For my recent project An Impossible Perspective, I was interested in understanding how artificial intelligence and machine learning might inform or impede our relationships with one another. So I used the imagery of plants as a metaphor to speculate on humanity’s future relationship with truth and labour. My goal with the project was to unveil the way that generative AI works and to demonstrate its functionality to the public. As an artist I am able to dedicate time to imagine alternatives to the way things function now. A way of doing that is filled with care and respect, and to bring this potential new way of being and doing to others to build that future together. To me, generative AI has many potential benefits. However, we—as artists, labourers, and members of society—must be involved in the development and use of such tools so that, as a community, we might have a chance to untangle ourselves from the pre-existing colonial and capitalist modalities within which these tools would otherwise operate and towards building something new.

Daphne: How does being an artist and educator inform the types of questions you’re investigating and your research methods?

Sanaz: For this project I devised my own unique research methodology and parameters which were all informed by an artistic framework. Using a set of AI tools, I worked with Millan Singh Khurana, a recent U of T graduate who studied both art and computer programming to create custom code with an intricate set of procedures aiming to mimic human intelligence. Through my artworks, I wanted to reveal to others how these technologies could present themselves as intelligent or creative. I used DALLE-2 and Midjourney to generate hundreds of images based on ideas such as the relationship between humankind and the natural world. Simultaneously, throughout a year I travelled to four different biomes to make over 10,000 photographs of plants. We then used both machine vision and natural language processing to train our own models from scratch, to be intelligent on artistic and poetic criteria such as the meaning of Realness, Artificiality, and Truthfulness.

Another important component of this project was the way I used my body to assemble the artworks. All my actions were predicated on the ordering and sequencing outcomes provided by the ML. My own labour, actions, and aesthetic choices were hence influenced by the AI software’s computational outcomes that directed the placement of each image. For example, in Red Flowers (Animate or Inanimate) pictured above, images on the top-left corner progress from ‘most deep fake’ to ‘least deep fake’ along horizontal rows, according to the score generated by the software. However, the surprise is that all of the images that the software was evaluating were deep fakes, with seventy-percent not being recognized by the AI. By sharing and showing this outcome to the viewers, the artwork demonstrates the extreme limitations of AI and the impossibility for it to make judgements on such essential distinctions.

Furthermore, the process of interacting with the AI is augmented by hand-cutting each image into a unique shape, inscribing the process with a physical and human creative act. This part of the art making was an extremely labour intensive and difficult process, resulting in hundreds of hours of work. Each of the thousands of paper forms were cut to resemble clothing pattern pieces and were then affixed with dressmaker pins to the artwork background. Here, I wanted to reference domestic labor and highlight the role of workers in our society—not just to consider their unfair working conditions, but also to address the lack of intellectual autonomy experienced by data workers and other labourers who have little to no control over the tasks they are required to perform in their places of work. Hence the process that I devised as an artist offers a space to experience the potential shift in autonomy that we may all face soon as a result of the use of AI. Revealing the tension between the goals of private enterprises versus our own aesthetic, intuitive and ethical desires.

Rafael: What’s your star sign?

Sanaz: Aries with a Gemini moon.

Follow Sanaz’s work with CLCF here.