Sex after Technology: The Rhetoric of Health Monitoring Apps and the Reversal of Roe v. Wade

ABSTRACT

The convergence of artificial intelligence technologies with the growth of Christo-fascist movements in the United States presents an alarming threat to women’s health, especially considering known privacy violations by the major players—all in the shadow of the US Supreme Court’s reversal of Roe v. Wade. These violations are ethotic; that is, they betray information that has been mined algorithmically to construct “user models,” bits and pieces of which are sold or otherwise circulated without true “user” consent or cooperation. Such models are best understood as algorithmic ethopoeia, mathematized representations of individuals charted as matrices of commodified categories for commercial trafficking, but also for politicians and law enforcement. Taking inspiration from abolitionist tools for resisting intersectional racism, and incorporating data feminism, we offer six categories of design heuristics to respect and maintain ethopoeic integrity, especially in the domain of women’s health in a post-Roe technological landscape, using a fundamental rhetorical concept to serve designers, as well as critics and activists.

KEYWORDS:

• Abortion rhetoric
• AI
• algorithmic ethopoeia
• menstruation rhetoric
• sexual surveillance

Acknowledgments

We thank the many scholars with whom we work at the University of Waterloo for their contributions to the research culture that makes this essay possible, singling out only a few of the more direct influences on our thought and approach: Danielle Deveau, Monique Kampherm, Marcel O’Gorman, and most especially Lai-Tze Fan, whose detailed and meticulous critique of an earlier draft and generous discussion of our ideas have improved the essay immeasurably. We also thank Joshua Trey Barnett, S. Scott Graham, and Zoltan Majdik for their attentive and encouraging guidance, as well as the anonymous reviewers for their thoughtful, productive readings of this work.

Disclosure Statement

No potential conflict of interest was reported by the author(s).

Notes

¹ «Lanzamos un programa para prevenir el embarazo adolescente utilizando inteligencia artificial … reconocidísima empresa de software del mundo.» (Página|12, 2018).

² «Con la tecnología vos podés prever cinco o seis años antes, con nombre, apellido y domicilio, cuál es la niña, futura adolescente, que está en un 86 por ciento predestinada a tener un embarazo adolescente.» (Página|12, 2018)

³ The case, brought by the American Civil Liberties Union, was settled in 2019 with Facebook promising to “eliminate age- and gender-based targeting as well as options for targeting associated with protected characteristics or groups” when selling ethotic data (Sherwin and Bhandari).

⁴ At least one other candidate term should be mentioned, the term data-consubstantiation, suggested by Hanford’s rich Burkean essay, “The Data Machine: Identification in the Age of Data Mining”; we set it aside largely because ethos commands a more familiar legacy in rhetoric than identification.

⁵ We are aware that, like many big terms in rhetoric, and lots of small ones too, ethopoeia is essentially contested. The most comprehensive account of the contestation, which lands close to where we are with the term, minus the algorithms, is Kampherm’s “Masks and Caricatures: Prosopopoeia, Ethopoeia, and the Effect of Social Media on Canadian Political Leaders’ Debates,” on which we have drawn liberally.

⁶ In software design, interoperability emphasizes the smooth integration and communication between different software and hardware systems. For businesses, the benefits include increased efficiency due to seamless data flow, the flexibility of system choices without compatibility concerns, fostering innovation through easier tech integration, and an enhanced user experience. For instance, interoperability between PillPack and One Medical would ensure streamlined data exchange and collaboration, ideally to optimize patient care and service delivery.

⁷ Matrix of domination she adapts from Patricia Hill Collins.

⁸ This work was first proposed in the senior author’s “Conversations Towards Practiced AI-HCI Heuristics” (Lubin).

⁹ These are 2003 numbers, reported by Project Prevention (Shatila et al. 34). The program is still in existence, with mild shifts in the racist and misogynist quotients of its paid clients (projectprevention.org/statistics/). Black “clients” now show up at a little under twice their expected numbers, as of 8 March 2023. Male “clients” are up to 6%. The payment has also increased slightly. It is now $300.

¹⁰ In her study of C.R.A.C.K.’s client ethopoeia (or what she calls their “ideological construction”), Erika Derkas found that “99% [of their characterizations] focus solely on gendered, class-based, and racialized pathology. Three main features defining all women targeted by CRACK are: (1) sexual deviance; (2) maternal aberration; and (3) individual defects” (”Don’t let”, 129).

¹¹ «es la excusa perfecta para que los activistas antimujeres y anti derechos sexuales y reproductivos declaren innecesarias las leyes de aborto.»

¹² «Las malas ideas disfrazadas de innovación se propagaron rápidamente.»

Additional information

Funding

This work was supported by the Social Sciences and Humanities Research Council of Canada under Grant [435-2022-0792]