Themed Research Seminar (Aesthetics, AI, Criticism, and Cultural Form) Session 2: Lightning Talks - AI mirrors, clones, ghosts, and cultural forms

Session 2: Lighting Talks - AI Mirrors, Clones, Ghosts, and Cultural Forms

Presented by: Dr Kiah Hawker; Dr Lisa Bode; Prof Jenna Ng; Prof Nic Carah

Date: Friday 1 May 2026
Time: 12-1pm
Location: Online via Zoom and in person at 09-738 (Level 7, Michie)

Format: Lightning Talks (10 mins)

Speaker 1: Kiah Hawker

Title: From AR (Augmented Reality) to AI: The Beauty Default and the Platformisation of Gendered Selfie-Making

Abstract: Beauty filters are among the most common filters algorithmically targeted to and applied by users on platforms like TikTok, Snapchat and Instagram - and yet the beautifying features coded into them are rarely chosen and increasingly obscured. This paper introduces the concept of the 'beauty default' to describe how beautification has become an unwritten expectation coded into nearly all filters, even those designed primarily for humour or play. Drawing on qualitative research with everyday users and makers of filters and critical analysis of the filters they use, I trace how the beauty default has been platformised over the past decade - where early AR filters largely overlaid static effects onto the face, contemporary filters like TikTok's Bold Glamour draw on machine learning and automation to edit the face in real time, moving from surface level tweaks to complete generative transformations. What makes the beauty default significant is not just that it reproduces hegemonic beauty norms, but that its obscurity and ubiquity produce uneven consequences. The same feature that deepens surveillance and the forensic gaze for feminine-presenting users is what allows masculine-presenting users to engage with beautification without social stigma, opening up new spaces for gender play. As filters shift toward generative AI, the beauty default becomes more deeply embedded in platform affordances and harder to see or challenge.

Bio: Kiah Hawker is a Lecturer in Digital Media in the School of Communication and Arts at the University of Queensland. Her research examines how emerging technologies shape digital culture, with a focus on AR and AI on the major platforms and how these tools shape practices of self-representation

Speaker 2: Lisa Bode

Title:  From “embodiment” to “agency”: shifting screen performer value in the era of AI clones and Tilly Norwoods

Abstract: This presentation draws from my current research historicizing the development of digital and (more recently) Gen AI replication of human bodies, faces, and voices in the commercial film and TV industry from the 1970s to the present. As one strand of that historicization, I look at how different Hollywood creative and craft labor groups have responded to developments in CG and Gen AI humans, and predictions from the late 1990s onwards that these figures would be used by producers to “replace” human performers. These responses accord with John T. Caldwell’s observation in Production Cultures (2008), that, during periods of significant industrial realignment and technological change, different craft labor communities make claims for the special aesthetic and cultural value of their field’s contribution to filmmaking. The greater the threat, he argues, the more insistent the claims to value and significance. We see this in all creative fields now in the era of generative AI. What is particularly interesting to me though is how the definitions of screen performance value from Hollywood’s screen acting communities and their industry allies have shifted across different moments of the digital era. While claims about acting’s value in the 2000s and 2010s emphasized the actor’s “embodied choices,” similar claims in the 2020s often talk more broadly about “agency” – without necessarily discussing the body. I will present evidence of this shift, and some thoughts on its causes and significance.

Bio: Lisa Bode is Senior Lecturer in Film and TV Studies in the School of Communication and Arts at UQ. Her research interests include technological change in filmmaking, screen performance, and the impact of digital bodies, deepfakes and generative AI on screen labor.

Speaker 3: Jenna Ng

Title: Digital Apparition Redux: From virtual humans to AI ghosts (and a bit of ABBA, as in the Swedish pop group) 

Abstract: This paper considers the phenomenon of virtual humans - from virtual influencers and virtual performers to AI re-animations of the dead - via media philosopher Vilém Flusser’s ‘digital apparition’ from his 1991 essay of the same title. As a virtual human, a person becomes a digital apparition of information density, with no fixed subjectivity with which to anchor experience, validate authenticity, grant consent etc. All its bases of interaction rest on its reality of a computable possibility of points. Flusser raised this issue as an anthropological question of the digital; we now need to think of them as questions of the existential: where the self is essentially a digital project, who are we? Where virtualization both reduces and expands the dimensionalities of reality, what is the story of the person/deceased, where is their authenticity? Virtuality thus point not only to issues of ethics and tech corporate power, but discombobulates the very basis of truth in the heuristic of realism: namely, how to evaluate what to believe and decide what is true based on what one sees and hears, and in a wider sense the very understanding of the world across science, reason, myth and magic.

Bio: Dr Jenna Ng is Professor and Head of School. She has research interests in digital media cultures; the creative technologies (interactive; AR; VR; XR); AI and algorithmic culture; machinima; and cinephilia.

Speaker 4: Nic Carah

Title: Thinking about automated media's structures of feeling: flow, tuning, vibes and slop 

Abstract: In this talk I draw on a larger book project emanating from a research project about Instagram I've been working with collaborators since the 2010s. In the earlier phases of this project we developed a purpose-built machine vision system that approximated the capacity of Instagram to cluster images together based on latent visual patterns. Using that tool we were able to explore both the visual cultures of Instagram and how machine vision systems made sense of them. We then worked with Instagram users who donated their Instagram archives to us, and through interviews that used scrollback, image sorting and machine vision approaches developed a form of intimate platform biography. I draw on this work to offer some concepts for thinking about the ordinary and everyday cultural forms generated by AI models as they sequence, augment and create images. I suggest that flow, tuning, vibes and slop are vernacular, technical and theoretical concepts that help us think about the emergent structure of feeling in media cultures increasingly organised by automated models. 

Bio: Nicholas Carah is Director of the Centre for Digital Cultures & Societies, Deputy Director of the Australian Internet Observatory and Associate Investigator in the ARC Centre of Excellence for Automated Decision-Making and Society.