The Psychology Behind AI - Investigating Biases in Crowdsourced Atomic Data Through a Sustainability Lens

Keywords:
Crowdsourcing, Sustainability, Bias, Machine Learning, Evolution

Abstract:
Progress towards accomplishing the Goals of the UN 2030 Agenda for Sustainable Development has been limited, and learning for environmental sustainability is fraught with challenges. Biases developed as an evolutionary mechanism for human survival; however, they can also hinder environmental sustainability in situations with unclear, uncertain outcomes. Crowdsourced databases, which are used to train Artificial Intelligence models via machine learning, often contain biases; therefore, this study presents a secondary data analysis of extracted data from the ATOMIC commonsense knowledge base. The extracted data consists of 1145 if-then statements linked to the environment. Binomial logistic regression was used to test whether a statement being biased or not predicts whether a statement is environmentally sustainable or unsustainable. The relationship between probability neglect and present bias and unsustainable statements was found to be statistically significant (p < 0.001), supporting the hypothesis that being biased corresponds to acting in an unsustainable way. These results are discussed through the lens of learning theory and the evolutionary psychology framework. Although limitations are present, this line of research has numerous global implications and has the potential to inform research both on Natural Language Processing (NLP) and educational programs targeting environmental cognitive biases. This has the potential to contribute to better educating humanity about the consequences of their actions and how to approach the environment in a more healthy, holistic and sustainable way.

CITATION:

IEEE format

Copy to clipboard

L. Leskur, “The Psychology Behind AI - Investigating Biases in Crowdsourced Atomic Data Through a Sustainability Lens,” in Sinteza 2026 - International Scientific Conference on Information Technology, Computer Science, and Data Science, Belgrade, Singidunum University, Serbia, 2026, pp. 557-564. doi:10.15308/Sinteza-2026-557-564

APA format

Copy to clipboard

Leskur, L. (2026). The Psychology Behind AI - Investigating Biases in Crowdsourced Atomic Data Through a Sustainability Lens. Paper presented at Sinteza 2026 - International Scientific Conference on Information Technology, Computer Science, and Data Science. doi:10.15308/Sinteza-2026-557-564

BibTeX format
Download

RefWorks Tagged format
Download