Ethical Training of Neural Networks: Datasets, Bias, and Privacy in Deepfake Detection
Ethical Training of Neural Networks: Datasets, Bias, and Privacy in Deepfake Detection
DOI:
https://doi.org/10.51473/rcmos.v1i2.2024.1871Keywords:
deepfakes; algorithmic ethics; dataset bias; digital privacy; informational sovereigntyAbstract
The automated detection of deepfakes through large-scale neural networks has become a civilizational necessity for protecting informational integrity in the digital environment. However, the most critical point in this architecture does not lie in the model itself, but in the ethical formation of the datasets used during training — which are frequently affected by structural biases related to race, gender, and geopolitics, as well as by potential violations of privacy and non-consensual biometric data extraction. This article develops an epistemologically rigorous analysis of the ethical dilemmas embedded in dataset construction for deepfake detection models, examining how flawed curation can reinforce algorithmic oppression, reproduce historical inequalities, and ultimately endanger democratic legitimacy. It concludes by proposing an ethical framework and engineering guidelines for responsible and auditable dataset governance, anchored in informational sovereignty and computational justice
Downloads
References
ARAL, SINAN. The hype machine: how social media disrupts our elections, our economy, and our health—and how we must adapt. New York: Currency, 2020.
BRUNDAGE, MILES; WHITTLESTONE, JESS. Governance of artificial intelligence: ethics, law and policy. Oxford: Oxford Internet Institute, 2020.
BUOLAMWINI, JOY; GEBRU, TIMNIT. Gender shades: intersectional accuracy disparities in commercial gender classification. Proceedings of Machine Learning Research, 2018.
CRAWFORD, KATE. Atlas of AI: power, politics, and the planetary costs of artificial intelligence. New Haven: Yale University Press, 2021. DOI: https://doi.org/10.12987/9780300252392
DOSHI-VELEZ, FINALE; KIM, BEEN. Towards a rigorous science of interpretable machine learning. arXiv, 2018.
FLORIDI, LUCIANO. The logic of information: a theory of philosophy as conceptual design. Oxford: Oxford University Press, 2020. DOI: https://doi.org/10.1093/oso/9780198833635.001.0001
GEBRU, TIMNIT et al. Datasheets for datasets. In: CONFERENCE ON FAIRNESS, ACCOUNTABILITY, AND TRANSPARENCY. 2018.
MILAN, STEFANIA; TRERÉ, EMILIANO. The rise of data colonialism: reclaiming digital sovereignty. Social Media + Society, 2019.
NATO STRATCOM. Deepfake detection and information integrity report. Brussels: NATO Strategic Communications Centre of Excellence, 2021.
VACCARI, CRISTIAN; CHADWICK, ANDREW. Deepfakes and disinformation: political campaigns in the AI age. Journal of Political Communication, 2020.
ZUBOFF, SHOSHANA. The age of surveillance capitalism. New York: PublicAffairs, 2019.
Downloads
Published
Issue
Section
Categories
License
Copyright (c) 2024 Matheus de Oliveira Pereira Paula (Autor)
This work is licensed under a Creative Commons Attribution 4.0 International License.
This work is licensed under the Creative Commons Attribution 4.0 International License (CC BY 4.0). This means you are free to:
- Share — copy and redistribute the material in any medium or format.
- Adapt — remix, transform, and build upon the material for any purpose, even commercially.
The use of this material is conditioned upon proper attribution to the original author(s), providing a link to the license, and indicating if changes were made. The license does not require permission from the author or the publisher as long as these conditions are met.
The Creative Commons license logo is displayed permanently in the footer of the journal.
The copyright of the manuscript can be retained by the authors without restrictions and can be requested at any time, even after publication in the journal.
How to Cite
