Social scoring in credit services in the context of artificial intelligence laws
Abstract
The use of artificial intelligence (AI) in the financial sector has garnered increasing attention in recent years, particularly in credit assessment and risk management. One of the innovative tools employed by financial institutions is social scoring, which utilises social media data to analyse a borrower’s profile.
This article aims to determine whether the use of social scoring in financial services complies with the new EU legislation. The article aims to demonstrate that the use of social scoring and artificial intelligence in financial services significantly violates the right to privacy, creating a high risk of unequal treatment and discrimination in the credit assessment process. The paper employs a dogmatic-legal approach, analysing provisions of Polish banking law and EU regulations concerning behavioural assessment using artificial intelligence in credit services. Subsequently, it examines current legal frameworks allowing for behavioural analysis in financial services, followed by an exploration of new regulations stemming from legislation on artificial intelligence. In conclusion, the fundamental values that limit social scoring were emphasised, and directions for the development of alternative forms of credit assessment were identified. Considering current regulations, the permissible use of automated creditworthiness analysis, based on financial and economic data, has been clearly distinguished from the prohibited practice of social scoring. The use of information from social media in the decision-making process is forbidden, and every automated decision must allow for an appeal to a human-conducted assessment.
Keywords:
artificial intelligence,, artificial intelligence act, social scoring, credit scoringReferences
Arroyo Amayuelas E., A Third Directive on Consumer Credit, European Review of Contract Law, vol. 20, no. 1, 2024, pp. 1-24., https://doi.org/10.1515/ercl-2024-2001 Google Scholar
Czech T. [w:] Kredyt konsumencki. Komentarz, wyd. III, Warszawa 2023, art. 9. Google Scholar
Durand D., Credit-Rating Formulae , Risk Elements in Consumer Instalment Financing, NBER, 1941 URL: http://www.nber.org/chapters/c9265. Google Scholar
Komisja Europejska. BIAŁA KSIĘGA w sprawie sztucznej inteligencji Europejskie podejście do doskonałości i zaufania PL, Bruksela, dnia 19.2.2020 r. COM(2020) 65 final. Google Scholar
Kosinski M., Khambatta P., Wang Y., Facial recognition technology and human raters can predict political orientation from images of expressionless faces even when controlling for demographics and self-presentation. American Psychologist 2024, https://doi.org/10.1037/AMP0001295 Google Scholar
Kostka G., China’s social credit systems and public opinion: Explaining high levels of approval, New Media & Society, 2019, 21(7), 1565-1593. https://doi.org/10.1177/1461444819826402 Google Scholar
Kotlarz M., Ocena zdolności kredytowej w prawie i orzecznictwie europejskim, PS 2023, nr 11-12, s. 124-138. Google Scholar
Kuchciński A., Ryzyko kredytowe w działalności banku, „Kwartalnik Naukowy Uczelni Vistula”, 2016, nr 2(48). Google Scholar
Lehmann M., Narzędzia wspomagające ocenę wiarygondości kredytowej podmiotór sektora MSP. Scoring Behawioralny i scoring zysku, „Zeszyty Studiów Doktoranckich/Uniwersytet Ekonomiczny w Poznaniu. Wydział Ekonomii”, nr46, 2009r.. Google Scholar
Liang F., Chen, Y., The making of “good” citizens: China's Social Credit Systems and infrastructures of social quantification, Policy & Internet, 2022, 14,114–135. https://doi.org/10.1002/poi3.291 Google Scholar
Mac Síthigh, D., Siems, M., The Chinese Social Credit System: A Model for Other Countries? Modern Law Review, 2019, 82(6), 1034-1071. https://doi.org/10.1111/1468-2230.12462 Google Scholar
Mednis A. [w:] Prawo bankowe. Komentarz, red. A. Mikos-Sitek, P. Zapadka, Warszawa 2022, art. 105(a). Google Scholar
Nierodka A., Część IV ZDOLNOŚĆ KREDYTOWA [w:] A. Nierodka, I. Heropolitańska, T. Zdziarski, Kredyty, pożyczki i gwarancje bankowe, Warszawa 2021. Google Scholar
Przanowski K., Credit scoring w erze big-data. Techniki modelowania z wykorzystaniem generatora losowych danych portfela consumer F.finance, Warszwa 2014. Google Scholar
License
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
