Trustworthy & accountable AI

Trustworthy & Accountable AI

AI through the lenses of experts on accountability and metrics.

Metrics play a central role in Artificial Intelligence (AI) applications. Employers use AI to filter job candidates and identify good future employee, so they measure extraversion, agreeableness, conscientiousness, neuroticism, and openness to ideas. Schools use AI to promote teachers, so they measure student test scores. Video streaming companies use AI to keep users engaged with the content, so they measure the number of hours spent watching videos. Unfortunately, using AI with such metrics can lead to undesirable consequences and reinforce discrimination in hiring practices, increase the cases of employee depression, and incentivize conspiracy theories.

On the other hand, metrics can push AI into precise and verifiable claims to which owners of AI can be held accountable. The design of metrics, the impact of metrics, and the use of metrics with the purpose of upholding accountability and ultimately increase trust has been a long endeavour of researchers in accounting and management information systems. In the Trustworthy and Accountable AI Lab, part of the Erasmus Center for Data Analytics (ECDA), AI is looked at through the lenses of experts on accountability and metrics.

Research Projects

Research by Iuliana Sandu is focused on:

Algorithms under control
Impact of AI on the accounting profession
Educating digital auditors

Relevant articles:

Sandu, M.I. & Koppius, O. (2019). Algorithms under control: An assertion-based framework for the audit of algorithms (working paper, November 2019).
Shmueli, G., & Koppius, O. R. (2011). Predictive analytics in information systems research. MIS quarterly, 553-572.
Dalla Via, N., Perego, P., & Van Rinsum, M. (2019). How accountability type influences information search processes and decision quality. Accounting, Organizations and Society, 75, 79-91.
Kramer, S., & Maas, V. S. (2016). Selective attention to performance measures and bias in subjective performance evaluations: an eye-tracking study. Available at SSRN 2457941.
Müller, M. A., Peter, C. D., & Urzúa I, F. (2020). Owner Exposure Through Firm Disclosure. Available at SSRN 3565224.
Stouthuysen, K., Teunis, I., Reusen, E., & Slabbinck, H. (2018). Initial trust and intentions to buy: The effect of vendor-specific guarantees, customer reviews and the role of online shopping experience☆. Electronic Commerce Research and Applications, 27, 23-38.
Vanhaverbeke, S., Balsmeier, B., & Doherr, T. (2019). Corporate Financial Transparency and Credit Ratings. Working paper.

Experts

PhD candidates

Tamara Thuis
Aljaž Sluga
Sebastian Stirnkorb
Albert Dongo
Anoek Leonieke Holthuijsen.

Download the presentation for more info!