The femininization of AI-powered voice assistants: Personification, anthropomorphism and discourse ideologies
Year: 2024
Type: article
Abstract: Intelligent Voice Assistants (IVAs) have been colonizing entire personal and professional online and offline scenarios (Statista, 2023). IVAs have been mainstreamed as female by default, through voices, avatars, colour palette, and conversational cues (Feine et al., 2020). Examples include Amazon Alexa, Apple Siri, Microsoft Cortana, and Google Assistant. Even though tech companies tend to justify this systematic feminization based on customers’ preferences (UNESCO, 2019), the ingrained harmful gender biases – which ultimately project images of subservient and obliging young females onto digital assistants – have been raising concerns in the last few years (West, Kraut, & Chew, 2019). These concerns pertain to the normalization of gendered, abusive, and toxic discourse practices that IVAs reproduce and, therefore, contribute to perpetuate.In this paper, a multimodal critical discourse approach (Machin, 2016), combined with feminist philosophy (Butler, 1988), and notions of ‘digital domesticity’ (Walker, 2020) will be applied to analyse examples of agents’ multimodal configuration that includes coded responses, as well as personification and anthropomorphic conversational cues (Bastiansen, Kroon, & Araujo, 2022). The analysis aims to uncover the companies’ hidden ideologies as they emerge from coded (i.e., pre-established) conversational practices (i.e., what IVAs are expected to say to engage users) in response to users’ prompts that gender, sexualize, and ultimately harass IVAs – a practice that hard-codes women and subservience. The paper seeks to advance understanding on the intersection of design interface of IVAs with reference to the ideological gendering of IVAs, actively pursued by designers to increase user engagement.
Source: Discourse Context & Media
Author: Maria Grazia Sindoni
Institution: University of Messina
Language: English
Subfield: Safety Research
Field: Social Sciences
Domain: Social Sciences
SDG: Gender equality
Open Access status: hybrid
Estimated APC paid: $3,540
OpenAlex ID: W4404396215