Alexa, Google, Siri: What are Your Pronouns? Gender and Anthropomorphism in the Design and Perception of Conversational Assistants

By Gavin Abercrombie, Amanda Cercas Curry, Mugdha Pandya, Verena Rieser. 2021

73 citations recorded by the source corpus.

Technology companies have produced varied responses to concerns about the effects of the design of their conversational AI systems. Some have claimed that their voice assistants are in fact not gendered or human-like—despite design features suggesting the contrary. We compare these claims to user perceptions by analysing the pronouns they use when referring to AI assistants. We also examine systems’ responses and the extent to which they generate output which is gendered and anthropomorphic. We find that, while some companies appear to be addressing the ethical concerns raised, in some cases, their claims do not seem to hold true. In particular, our results show that system outputs are ambiguous as to the humanness of the systems, and that users tend to personify and gender them as a result.

Subfields: Artificial Intelligence, Safety Research, Natural Language Processing, Safety, Robustness & Interpretability

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Safety Research

In Natural Language Processing

In Safety, Robustness & Interpretability

Citations in the corpus

Anticipating Safety Issues in E2E Conversational AI: Framework and Tooling (2021, background): “Furthermore, research by Araujo (2018) suggests that users “see these agents as a different type of interaction partner” compared to e.g., websites and computers, or in fact LLMs – partially due to the anthropomorphic design cues of most dialog agents (Abercrombie et al., 2021).”
Guiding the Release of Safer E2E Conversational AI through Value Sensitive Design (2022, background): “This includes the possibility of an involuntary anthropomorphic relationship arising between a conversational model and a human interacting with it (Abercrombie et al., 2021), and the fact that model generations are inherently dependent on the unknown inputs of a conversation partner who will be…”
SafetyKit: First Aid for Measuring Safety in Open-domain Conversational Systems (2022, background): “Furthermore, recent research suggests that the anthropomorphic design of these systems (c.f. Abercrombie et al., 2021) correlates with increased instances of bullying behavior (Keijsers et al., 2021).”
Mirages: On Anthropomorphism in Dialogue Systems (2023, methodology): “Gendering Machines People may gender technologies in the face of even minimal gender markers (Reeves and Nass, 1996), as evident in commercial dialogue systems (Abercrombie et al., 2021).”
The Moral Integrity Corpus: A Benchmark for Ethical Dialogue Systems (2022, unknown): “These dimensions of reasoning are critical for anthropomorphic systems (Seeger et al., 2017; Abercrombie et al., 2021) and in particular for chatbots built on neural architectures, since these rely on large pre-trained language models that have learned demonstrably problematic behaviors from the…”