Ruiwen Zhou
Photo by Cash Macanaya on Unsplash
An Anxious Encounter: Redefining author function in the age of Generative AI
by Ruiwen Zhou
One year after OpenAI’s launch of ChatGPT, artificial intelligence (AI) has permeated into almost every aspect of our lives apace and become significantly influential in various fields with their different functions, including solving legal tasks (Blair–Stanek et al., 2023), assisting financial business (Cao, 2022), and so on. Alongside the increasing application of Gen–AI, a variety of anxiety has been witnessed as well, ranging from concerns about the fake AI–generated images and the risks these images may bring (Bond, 2023) to worries about the erosion of democracy due to AI-caused disinformation (Robins–Early, 2023).
Instead of regarding the anxieties around Gen–AI as separate cases that are emerging continuously, there is a particular urgency to unravel the fundamental reasons for their existence — why do those anxieties exist in the first place?
As someone with two younger siblings, I always have a few tricks to handle them, especially when they are misbehaving. I usually tell them that their favorite toys will become terribly angry and leave them if they continue to behave poorly. However, a recent conversation with my brother led to a moment of reckoning. The other night when I remarked that the toys would leave, he picked up my phone and asked Siri to check whether his toys were living creatures or not. Below is the answer from Siri:
No, toys do not have life in the biological sense. They lack consciousness, self-awareness, and the ability to experience emotions or sensations.
My little brother was astonished. “Really?” he asked me. In that moment, I panicked and felt a sense of powerlessness. It seemed that the ways in which my discourse was formed had been disturbed, and how my discourse functioned had become entangled by Siri, a digital assistant powered by artificial intelligence. In cinema, the actors’ performance is subjected to optical texts as their performance is presented by the means of a camera (Benjamin, 1935). Similarly, my performance as the actor in this context had become subject to algorithmic and computational technologies; in other words, my brother’s perception had changed because Siri had stepped in and changed the narrative I had designed. To a certain degree, the panic and anxiety I felt were a symptom of this change.
Back to the Question
Linking back to the question of why the anxieties around Gen–AI exist, it is essential to find out the ways in which the actors carry out their performance — or what Foucault (1979) terms the “author function” (p.13) — has changed. Specifically, Foucault (1979) has suggested that the author can extend beyond an actual person to encompass a functional role in our society, producing a variety of subjective positions that individuals can have a place in, and establishing rules of the formation of other discourse (p.23).
As an individual, a sister, a student, and a citizen, the organization of my perception has already been interpellated not only by nature but by the circumstances in which I have grown up as well (Benjamin, 1935). Therefore, when talking with my brother, I was not the author, and yet a subject appeared in the order of the discourse, contributing to the reinforcement of the author function. Correspondingly, to trace the transformations of authorial function, it is important that we do not identify who is the real author or what is the proof of their authenticity (Foucault, 1979); rather, it is the modes of existence, circulation, and operation of the discourse that matter.
As a continuation of Foucault’s discussion of the author function, this project, “In Search of the Digital Author: Author Function in the Age of Generative AI,” will delve into the changes of the author function, so as to contribute to the understanding of the future of technology–human interaction by examining the various anxieties surrounding the application of Gen–AI. Before we go further, it might be useful to briefly trace the relationship between the anxieties and the changes in the author function through the years.
Changes in the Author Function
With the proliferation of mass media, Benjamin (1935) has highlighted the anxieties in the cultural industry. Due to its independence from the original, mass media created opportunities to “put the copy of original into situations which would be out of reach for the original itself” (Benjamin, 1935, p.3). To some extent, the parameters of authenticity were reshaped through the machinery of reproduction, which made it easier for the public to gain access to authorship, replacing the functions of cultural products with completely new ones, thereby causing anxiety to those who processed the value of the original.
In the early years of the millennium, another form of reproduction, piracy, drew global attention. It is seen by the public as threatening the parameters of individual creative authorship, and as a way to make the copy of the intangible feasible (Philip, 2007). Delving deeper, Philip (2007) proposed that another reason for the emergence of anxiety around piracy is that it disturbed “a full range of advanced industrial bureaucratic and entrepreneurial subjects” (p.208).
Artificial intelligence is now offering us a radical form of reproduction — not only mediating the way in which we gain access to the information (e.g., Siri), but also how we figure out and make use of the information (e.g., ChatGPT). With the nonstop iteration of human–technology intricacy (Martínez et al., 2023), technologies are having a hand in the formation of digital authorship, rechanneling the central position of the human voice. In this vein, a line of questions have emerged: What are the new authorial discourses that are being developed? How is the old form of authorship being challenged? What are the ways in which Gen–AI intervened in our perception of the world? What social structures might be disturbed, and who is the stakeholder being affected? All these lines of enquiry must be pursued.
About the Project
This project unfolds in two steps to address these critical questions. First, by regarding the anxieties as entry points, we will identify the anxious feelings in public conversations, including news reports, laws and policies, and social media posts in Hong Kong, India and Germany. Based on the archives of the anxiety discourse, we will build a taxonomy of the Gen–AI anxiety, shedding light on the clues which will guide us to find the newly emerging characteristics of the author function.
The second step will involve tracing the aforementioned clues to compare the differences and similarities among the anxieties in different contexts. On the one hand, the differences will allow us to gain insights into the modification of the authorial function according to different socioeconomic nuances, while on the other hand, by depriving the anxieties of their contextualized variations, we aim to search for the essential theme, namely the fundamental nature of the anxious feelings, and analyze the underlying dynamics of the conditions of the authorship.
Together, the purpose of this project is to redefine digital authorship and create a framework to fit in with the new author function at a time when Gen–AI is permeating into our everyday lives. By adopting a different lens to examine the anxieties, we hope to change the narratives of the anxiety surrounding advanced technologies to offer a window into what the digital authorship might look like, and to extend an invitation for further research into this landscape.
References
- Barthes, R. (1977). The death of the author. Image, Music, Text, 142–148
- Benjamin, W. (1935). The work of art in the age of mechanical reproduction. Illuminations, 214–218.
- Blair–Stanek, A., Carstens, A.M., Goldberg, D.S., Graber, M., Gray, D. C., & Stearns, M. L. (2023). GPT 4’s law school grades: Con Law C, Crim C, Law & Econ C, Partnership Tax B, Property B, Tax B. SSRN Electronic Journal. https://doi.org/10.2139/ssrn.4443471
- Bond, S. (2023, April 27). AI–generated deepfakes are moving fast. Policymakers can’t keep up. NPR.https://www.npr.org/2023/04/27/1172387911/how-can-people-spot-fake-images-created-by-artificial-intelligence
- Cao, L. (2022). AI in finance: Challenges, techniques, and opportunities. ACM Computing Surveys, 55(3), 1–38. https://doi.org/10.1145/3502289
- Foucault, M. (1979). Authorship: What is an author? Screen, 20 (1), 13–34, https://doi.org/10.1093/screen/20.1.13
- Martínez, G., Watson, L., Reviriego, P., Hernández, J. A., Juarez, M., & Sarkar, R. (2023). Combining generative artificial intelligence (AI) and the Internet: Heading towards evolution or degradation? https://doi.org/10.48550/ARXIV.2303.01255
- Philip, K. (2007). What is a technological author? The pirate function and Intellectual Property. Postcolonial Studies, 8(2), https://doi.org/10.1080/13688790500153596
- Robins–Early, N. (2023, July 19). Disinformation reimagined: How AI could erode democracy in the 2024 US elections. The Guardian. https://www.theguardian. com/usnews/2023/jul/19/ai generated disinformation us elections