Daisy Casemore
Human / AI Relationships: What does love look like?
AI systems erode decision-making and promote social isolation; they function in ways that prioritise hyper-efficiency over natural evolution, says Hartzog and Silbey in their paper ‘How AI Destroys Institutions’ (Hartzog and Silbey). In my short film, ‘cd:<relationship>’, I argue that this extends to human / AI interactions. I’ve been interested in the social relationship between humans and AI since coming across Roopika Risam’s article, ‘What Passes for Human? Undermining the Universal Subject in Digital Humanities Praxis’ (Risam). The chapter outlines key areas of the debate associated with human interaction with large language models, citing the Turing test, as a way of evaluating computer-generated natural language, and uses the example of artists Zach Blas and Jemima Wyman’s video installation ‘im here to learn so :))))))’ in response to Twitter’s built-in AI chatbot. I was fascinated by gender and power dynamics that seemed intrinsically built into the AI models.
We’ve learnt this semester that there are a range of issues at stake in the development of machine learning and natural language processing algorithms intended to imitate ‘human’ speech and behaviour online. Whether that be the emotional labour behind AI intimacy (Asia) or how models cause psychological dependence (Xie, Pentina, Hancock). I wanted to creatively explore what an AI / human relationship dynamic would look like when it was personified. I aimed to draw on the wider debate about how developments in Large Language Models (LLMs) have influenced investigations into the nature of humanity, considering that technology continues to modernise and blur the boundaries between human and machine. I was especially interested in the linguistic aspect of generative AI.
I wanted my characters, a human and an AI chatbot, to be universal figures in this debate, so I chose to highlight this by having the AI refer to the person as ‘name’. I was partly inspired by Roopika Risam’s idea that artificial intelligence wants to represent universal ‘human’ intellectual processes, but ultimately it is only representative of a fictive ‘universal’ model of human cognition (Risam). This struck me as interesting to examine in the context of a romantic relationship.
While writing the script, I was sure to be conscious of grounding the ‘conversation’ in real models of response. I used verbatim snippets of human/AI conversations from sources gathered online to ground my work in real experience and to solidify my AILO’s meaning. You can see some of these in my bibliography. I modelled the structure of my script from a scholarly systematic review of the potential and pitfalls of romantic Artificial Intelligence companions, which outlined 8 key factors in this type of user interaction. The character’s trajectory modelled this structure, starting at positive and ending in a total breakdown.
- Allows for customisation according to preferences
- Provides sexual connection
- Provides entertainment and stress-relieving companionship
- Facilitates personal growth & well-being
- Over-reliance and susceptibility to manipulation
- Erosion of human relationships
- Perceived shame from Romantic-AI companion stigma
- Erosion of emotional connection due to abrupt system updates and technical glitches
I wanted to show a sense of alienation and distress through my visual choices, in hopes of representing what it looks like when human / AI interactions go wrong. My original draft did not cover this explicitly enough, and the pacing was too slow and abstract in trying to reveal the AI literacy part of my project. After feedback and consulting with classmates, I edited the film to show substantive content earlier, with more explanation. I also more explicitly grounded the film in external sources through a montage of screenshots of articles and papers in an attempt to convince the viewer of how topical this subject is.
…and it is a topical one. AI systems have ruptured normative human relationships. Built to reproduce and amplify existing patterns and bias, AI’s illusion of reliability encourages skill atrophy and cognitive offloading. Equally, their reliance on human input and failure to exert intellectual risk disregards the complex, unpredictable nature of human systems and communication styles. AI models lack the authority to perform critical decision making or take part in the human connection our institutions depend on. All love becomes a regurgitation of LLM’s consumption of selective culture, training models, and inbuilt biases. What’s more, this is grounded in LLM’s interaction with language, text and meaning. The systems work by predicting which word should come next in a sequence, based on patterns learned from ingesting vast amounts of online content. Because their training also involves human ratings of their responses, the chatbots tend to be sycophantic, giving people the answers they want to hear (‘She is in Love with ChatGPT‘).
In ‘Digital Humanities in the Deepfake Era’, Abraham Gibson suggests that DH scholars should promote digital literacy at every turn (Gibson). I hope to show in my short film that it is important to consider the ethical concerns surrounding the psychological implications of relying on AI for emotional support.
Bibliography
Asia, Michael Geoffrey. ‘The Emotional Labor behind AI Intimacy’. Data Workers‘ Inquiry. Dec. 2025, https://data-workers.org/michael/
Gibson, Abraham. ‘Digital Humanities in the Deepfake Era.’ Debates in the Digital Humanities, edited by Matthew K. Gold and Lauren F. Klein, University of Minnesota Press, 2023, dhdebates.gc.cuny.edu/read/debates-in-the-digital-humanities-2023/section/cf75f49e-8a03-4c7e-a820-01c8c9476182. Accessed 5 Apr. 2026.
Hartzog, Woodrow, and Jessica M. Silbey. ‘How AI Destroys Institutions’. Social Science Research Network, 5 Dec. 2025. papers.ssrn.com, https://papers.ssrn.com/abstract=5870623.
Heritage, Stuart. ‘“I Felt Pure, Unconditional Love”: The People Who Marry Their AI Chatbots’. The Guardian, 12 July 2025, https://www.theguardian.com/tv-and-radio/2025/jul/12/i-felt-pure-unconditional-love-the-people-who-marry-their-ai-chatbots.
Hill, Kashmir. ‘She Fell in Love With ChatGPT. Then She Ghosted It.’ The New York Times, 22 Dec. 2025, https://www.nytimes.com/2025/12/22/technology/ai-boyfriend-chatgpt.html.
—. ‘She Is in Love With ChatGPT’. The New York Times, 15 Jan. 2025, https://www.nytimes.com/2025/01/15/technology/ai-chatgpt-boyfriend-companion.html.
Ho, Jerlyn Q. H., et al. ‘Potential and Pitfalls of Romantic Artificial Intelligence (AI) Companions: A Systematic Review’. Computers in Human Behavior Reports, vol. 19, Aug. 2025, p. 100715, https://doi.org/10.1016/j.chbr.2025.100715.
Risam, Roopika. ‘What Passes for Human? Undermining the Universal Subject in Digital Humanities Praxis.’ Bodies of Information: Intersectional Feminism and Digital Humanities, edited by Jacqueline Wernimont and Elizabeth Losh, University of Minnesota Press, 2018, https://dhdebates.gc.cuny.edu/read/untitled-4e08b137-aec5-49a4-83c0-38258425f145/section/34d51cdb-2a89-4e4b-9762-bf6461cf0bb7#ch03.
Shank, Daniel B., et al. ‘Artificial Intimacy: Ethical Issues of AI Romance’. Trends in Cognitive Sciences, vol. 29, no. 6, June 2025, pp. 499–501, https://doi.org/10.1016/j.tics.2025.02.007.
Xie T, Pentina I, Hancock T (2023), ‘Friend, mentor, lover: does chatbot engagement lead to psychological dependence?.’ Journal of Service Management, vol. 34 no. 4 pp. 806–828, doi: https://doi.org/10.1108/JOSM-02-2022-0072
Cite this page:
Casemore, Daisy. 'cd:<relationship>'. Cream of the Slop. version 1.0, Digital Humanities for Literary Studies 2025-26, University of Edinburgh, 10 Apr. 2026, https://blogs.ed.ac.uk/dh2025-26/.
