The Spanish Telecommunication Sector and the Development of AI Products from a Gender Perspective

Projektinformationen

TELAIGEN

ID Finanzhilfevereinbarung: 101028792

DOI

10.3030/101028792

Das Projekt endete am 28 Februar 2023

EK-Unterschriftsdatum 27 April 2021

Startdatum 1 Februar 2022

Enddatum 31 Januar 2024

Finanziert unter

• EXCELLENT SCIENCE - Marie Skłodowska-Curie Actions

Gesamtkosten

€ 160 932,48

EU-Beitrag

€ 160 932,48

Koordiniert durch

UNIVERSIDAD CARLOS III DE MADRID
Spain

Periodic Reporting for period 1 - TELAIGEN (The Spanish Telecommunication Sector and the Development of AI Products from a Gender Perspective)

Berichtszeitraum: 2022-02-01 bis 2024-01-31

Zusammenfassung vom Kontext und den Gesamtzielen des Projekts

Given the rising relevance of AI-driven technology and the growing concerns over its reproduction of a gender bias, my research aims to respond to these tensions empirically through exploring the gender bias within the telecommunications industry in Spain. The chief objective of TELAIGEN is to develop a multi-methodological study of the Spanish telecommunications industry on a multi-level basis. On one level, the goal of TELAIGEN is to create a holistic picture of the Spanish telecommunications industry by ascertaining its gender gap and identifying where this gap is more prevalent within the sector (i.e. in AI research teams of AI developing teams) using quantitative-means. On another level, the goal of TELAIGEN is to determine the incorporation of a gender perspective in the sector that shapes the development of AI technologies through the implementation of qualitative research methods. This goal will be realised carrying out a case-study of the company Telefónica, Investigación y Desarrollo (TID), a subsidiary enterprise, part of the larger Telefónica Group, a private Spanish multinational telecommunication enterprise headquartered in Madrid, Spain. By focusing on the particular situation of one Spanish company, TELAIGEN will be able to produce a comprehensive report that effectively identifies the necessary adjustments and measures to be taken by TID to reduce the gender gap in their teams and the gender bias in their AI products.

Arbeit, die ab Beginn des Projekts bis zum Ende des durch den Bericht erfassten Berichtszeitraums geleistet wurde, und die wichtigsten bis dahin erzielten Ergebnisse

The Fellow has carried out critical work during the project's implementation. This work is divided between four work packages that cover dissemination and exploitation (publications and conference presentation), data collection (both primary and secondary, as well as quantitative and qualitative).
The publications undertaken by the Fellow are the following:
DÍAZ-FERNÁNDEZ, S. (2023) Engaging critically with AI: On biases, human-machine intra-action and gendered (dys)affordances. En Desafíos éticos, jurídicos y tecnológicos del avance digital. IUSTEL (Pending publication).
DÍAZ-FERNÁNDEZ, S., SAINZ DE BARANDA, C. (2023) Una aproximación a los discursos sobre IA ética en la industria Española. (in preparation)

The conference presentations undertaken by the Fellow are the following:
DÍAZ-FERNÁNDEZ, S. & SANDOVAL-MARTÍN, T. (2022). Análisis de los planes de igualdad del sector de las telecomunicaciones: de la brecha a los sesgos de género en la inteligencia artificial. II Congreso Internacional Tecnologias I+D+I para la igualdad: soluciones, perspectias y retos. (CITI+D+I) Carlos III University.
DÍAZ-FERNÁNDEZ, S. (2022) ‘Engaging critically with AI: On biases, human-machine intra-action and gendered (dys)affordances’. Congress of Artificial Intelligence & Gender: gaps, biases and new forms of inequality. (CAIG) Salamanca Univeristy. Universidad de Salamanca. Spain.
DÍAZ-FERNÁNDEZ, S. (2022) ‘Women like to be controlled”: exploring the chatbot misogyny of Replika and the limitations to women’s digital intimacy’. II International Congress of Digital Feminism. Women’s rights in the era of the Internet. (ICDF) Granada University. Spain.
DÍAZ-FERNÁNDEZ, S. (2022) ‘Postfeminist politics? Exploring equality efforts in the Spanish digital sector in a post-Covid context’. Società Italiana di Scienza Politica Conference 2022. (SISPC) Società Italiana di Scienza Politica. Italy.
DÍAZ-FERNÁNDEZ, S. (2023) Intimacy, recognition and algorithmic misogyny: the case of Replika. Algorithms for Her? University of Sheffield.
DÍAZ-FERNÁNDEZ, S. (2023) ‘Make AI political again! Exploring the framing of AI Ethics, bias and depoliticisation’ XX ISA World Congress of Sociology. 2023. Australia. (upcoming)
DÍAZ-FERNÁNDEZ, S. (2023) ‘Intimacy, recognition and algorithmic misogyny: the case of Replika’. XX ISA World Congress of Sociology. 2023. Australia. (upcoming).
DÍAZ-FERNÁNDEZ, S. (2023) ‘Who frames AI? Exploring AI developers' understanding of bias, fairness and their socio-political implications’. 20th Conference of Europeanist Studies. Council of European Studies. Reykjavik (upcoming).

The Fellow has also completed a final report upon completing the secondment therefore meeting the major outcome of the project.

Fortschritte, die über den aktuellen Stand der Technik hinausgehen und voraussichtliche potenzielle Auswirkungen (einschließlich der bis dato erzielten sozioökonomischen Auswirkungen und weiter gefassten gesellschaftlichen Auswirkungen des Projekts)

The work conducted illuminates the need for a clear gender perspective in the implementation of ethical frameworks in the development of AI technologies by key actors in the industry. Through the investigation, the Fellow determines current industry players within the telecommunications sector are pushing for ethical efforts that due to the lack of integration of a feminist and gender view are lacking with regards to developing effective ethical frameworks. This may result in the continuous reproduction of gender biases. In addition, current development of ethical principles to shape the AI innovation are also lacking a social science perspective, which consequently leads to the framing of the bias problem as a technical glitch to be fixed via technological means only. In this regard, the Fellow advises the incorporation of social science experts (feminist scholars and activists, sociologists, philosophers and anthropologists) to teams dedicated to the developing of AI technologies.

screenshot-2023-03-24-at-17-38-38.png