• Skip to main content
  • Skip to primary sidebar
  • Skip to footer
  • Core Principles of Responsible AI
    • Accountability
    • Fairness
    • Privacy
    • Safety and Security
    • Sustainability
    • Transparency
  • Special Topics
    • AI in Industry
    • Ethical Implications
    • Human-Centered Design
    • Regulatory Landscape
    • Technical Methods
  • Living Dictionary
  • State of AI Ethics
  • AI Ethics Brief
  • 🇫🇷
Montreal AI Ethics Institute

Montreal AI Ethics Institute

Democratizing AI ethics literacy

Theorizing Femininity in AI: a Framework for Undoing Technology’s Gender Troubles (Research Summary)

February 1, 2021

🔬 Research summary contributed by Alexandrine Royer, our Educational Program Manager.

[Link to original paper + authors at the bottom]


Overview: The predominance of feminized voice assistants points to AI’s tendency to naturalise gender divisions. This paper draws on the science fiction narrative of Her and Tomorrow’s Eve to offer a critical understanding of how femininity serves as a means of domesticating AI all the while reproducing gender relations.


For anyone keeping a close eye on technological development, the statement that AI has a gender problem in its conceptualization, design, and output comes as no surprise. AI automates the existing social status quo, and its developers deploy consumer products that reproduce deeply engrained gender divisions and roles. In 2001, Microsoft’s Halo: Combat Evolved game introduced the character of Cortana, an AI-powered female voice and occasional sexy cybernetic hologram, made to assist main players. By no coincidence, Cortana would later become the name of Microsoft’s feminized and subservient artificially intelligent voice-interface, in the likes of Apple’s Siri and Amazon’s Alexa. The way gender is imagined will naturally feed into how it is materialized; as Sutko asserts, “technological representation [sic] intersects with technological design.” 

Responding to what he terms “technofeminist calls to unpack the gendered politics of technologies,” Sutko offers a critical analysis of how ideas of femininity continue to be reproduced in technologies across three dimensions: docile labour, replaceable embodiment, and artificial intelligence. The author draws on science fiction narratives to underscore how femininity serves as a mode of domesticating new technology, with technology, in turn, materializing these ideas of gender relations. For Sutko, “the dreams and designs of AI (re)produce discursive formations rooted in the subjugation of others, particularly women”. Combining insights from media archaeology, critical theory, gender studies, and media theory, Sutko delineates how gender divisions are sustained and naturalized through technology.  

Fictional Characters – Real Illustrations 

Novels, movies, video games, and other forms of media provide a framework to evaluate the narratives, tropes, and underlying gender patterns in the depiction and, often idealized views of, the creation of artificial intelligence. A recurrent plotline in science fiction literature and cinema revolves around masculine relationships with feminized technologies. Spike Jonze’s 2013 Her depicts a lonely middle-aged divorcee who falls in love and begins a sexual relationship with his seductively voiced virtual assistant. A century back, the misogynistic Victorian novel Tomorrow’s Eve centers on a scientist who invents a female android for his lovelorn aristocratic friend; the aristocrat seeks to replace his shallow wife with an artificial ‘ideal’ woman. 

Both stories are told at a turning point of new technology, Eve during the introduction of technical media (e.g. the gramophones, photography, cinematography, etc.) and Her during the introduction of computational media (e.g. smartphones, virtual assistants, etc.). As Sukto underlines, “cultural artifacts like Eve and Her are useful to think with because they help us identify which structures of feeling are being dismantled or reinforced.” Adding that, despite being more than a hundred years apart, “both stories envision using contemporary communication technologies to create an ideal female lover and associate femininity with new technology.” In Her and Eve, female technologies are made to produce affective labour docilely and replace real-life ‘dysfunctional women’, yet their artificiality is also a threat to the men who own them- reminding them of their very ‘unnaturalness’. 

Feminized Labour, Embodiment and Intelligence

By arranging our calendars, making calls, and sending reminders, voice interfaces are designed to provide stereotypically gendered labour. As argued by Sutko, “gendering the interface-whether Sam or Siri- essentializes labour divisions and female bodies/minds as docile, disciplined to respond to others’ needs”. This pattern reinforces the view that men master technology and women are subordinate to it. Yet, Sutko cautions that this division is muddled by our daily interactions with technology, where Siri is often the one programming us and not the other way around.  Speaking from his own experience, Sutko notes, “I have been trained by Siri to enunciate properly, to ask only certain kinds of questions, in a particular way, to get a useful response.” 

Along with changing or reaffirming conceptions of labour, virtual assistants and androids complicate our understanding of the division between mind and body. For Sutko, “both Eve and Her erase the politics of embodiment by reducing embodiment to a state of mind.” Sutko points out how both stories reinforce a view of the unknowable inner workings of feminized technology –“who knows why they act that way” – all the while depicting them as machines who are programmed to care, nurture and comply. This tension replicates familiar tropes surrounding femininity and the characterization of women. 

As Sutko summarizes, “essentialized categories of gender, nature, culture and technology are replicated in our fantasises of artificial intelligence without introspection and may be reproduced or rearticulated depending on the design of such AI”. 

Though our fears and fantasies over intelligent machines often eclipse this fact, AI is an artificial imitation of human intelligence – it is an attempt to create something equal and less than human. In Sutko’s view, there is a willingness to have AI systems respond to all our desires without affording them any of their own. Ideal AI must be “caring, nurturing, responsive, attentive, helpful but not willful, smart but not overly so, replaceable, customizable, available.” Computer scientists know that any AI system is unlikely to meet all these requirements soon. Nevertheless, the spread of voice interfaces, and popular fantasized depictions, will impact human cognition and communication. Simulations are also not just descriptive but prescriptive. 

Going Forward – Prescriptions for Future AI

As technological development advances, scientists will continue their endeavours to mirror human cognition and behaviour in machine-shaped forms- with what it means to be human remaining a shifting category. Sutko argues that society must ‘take seriously’ these fictionalized accounts of AI: “If we consider the expectations for AI seriously as anticipatory models of our own future, then we open up political possibilities for an ethical obligation towards AI”. 

Both Her and Eve also point to the changes in human behaviour as the leading characters co-evolve alongside their technological companions. As Sutko underscores, human behaviour will be modified by closer technological interaction- “we become machine-shaped”, turning the Turning test on its head. His prescriptions for future interventions include introducing concepts of politeness, consent, and respect in “the way AI addresses us” but also “the way we are required to address AI.” Other suggestions include equipping AI with gender-neutral tones or allowing the AI to be modulated according to the user’s preference. While Sutko recognizes that these solutions do not redress the long-standing structural inequalities that have led to feminized AI, he argues that “they are a place to begin unpacking the black-boxed intersectionalities of gender, bodies, labour, and technology”. 

With AI companions continuing to pervade multiple spheres of human activity, a thorough and critical examination of its development beyond the rational/technical and into the heuristic will add some welcomed nuances to our understandings of the intersection of gender and AI. 


Original paper by Daniel M. Sutko: https://doi.org/10.1080/09502386.2019.1671469

Want quick summaries of the latest research & reporting in AI ethics delivered to your inbox? Subscribe to the AI Ethics Brief. We publish bi-weekly.

Primary Sidebar

🔍 SEARCH

Spotlight

AI Policy Corner: The Kenya National AI Strategy

AI Policy Corner: New York City Local Law 144

Canada’s Minister of AI and Digital Innovation is a Historic First. Here’s What We Recommend.

Am I Literate? Redefining Literacy in the Age of Artificial Intelligence

AI Policy Corner: The Texas Responsible AI Governance Act

related posts

  • Research summary: Snapshot Series: Facial Recognition Technology

    Research summary: Snapshot Series: Facial Recognition Technology

  • Friend or foe? Exploring the implications of large language models on the science system

    Friend or foe? Exploring the implications of large language models on the science system

  • Achieving Fairness at No Utility Cost via Data Reweighing with Influence

    Achieving Fairness at No Utility Cost via Data Reweighing with Influence

  • Prompt Middleware: Helping Non-Experts Engage with Generative AI

    Prompt Middleware: Helping Non-Experts Engage with Generative AI

  • What lies behind AGI: ethical concerns related to LLMs

    What lies behind AGI: ethical concerns related to LLMs

  • On Measuring Fairness in Generative Modelling (NeurIPS 2023)

    On Measuring Fairness in Generative Modelling (NeurIPS 2023)

  • RAIN Africa and MAIEI on The Future of Responsible AI in Africa (Public Consultation Summary)

    RAIN Africa and MAIEI on The Future of Responsible AI in Africa (Public Consultation Summary)

  • The Role of Arts in Shaping AI Ethics

    The Role of Arts in Shaping AI Ethics

  • Listen to What They Say: Better Understand and Detect Online Misinformation with User Feedback

    Listen to What They Say: Better Understand and Detect Online Misinformation with User Feedback

  • Down the Toxicity Rabbit Hole: Investigating PaLM 2 Guardrails

    Down the Toxicity Rabbit Hole: Investigating PaLM 2 Guardrails

Partners

  •  
    U.S. Artificial Intelligence Safety Institute Consortium (AISIC) at NIST

  • Partnership on AI

  • The LF AI & Data Foundation

  • The AI Alliance

Footer

Categories


• Blog
• Research Summaries
• Columns
• Core Principles of Responsible AI
• Special Topics

Signature Content


• The State Of AI Ethics

• The Living Dictionary

• The AI Ethics Brief

Learn More


• About

• Open Access Policy

• Contributions Policy

• Editorial Stance on AI Tools

• Press

• Donate

• Contact

The AI Ethics Brief (bi-weekly newsletter)

About Us


Founded in 2018, the Montreal AI Ethics Institute (MAIEI) is an international non-profit organization equipping citizens concerned about artificial intelligence and its impact on society to take action.


Archive

  • © MONTREAL AI ETHICS INSTITUTE. All rights reserved 2024.
  • This work is licensed under a Creative Commons Attribution 4.0 International License.
  • Learn more about our open access policy here.
  • Creative Commons License

    Save hours of work and stay on top of Responsible AI research and reporting with our bi-weekly email newsletter.