State of use of AI tax systems
The first mention of the use of AI systems by the tax administration of the Republic of Slovenia dates back to 2017.
What functions are performed with AI?
Based on publicly available data, the only function performed by a tax machine-learning algorithm is:
- External risk-management (risk-scoring): the tool predicts the risks of tax non-compliance associated with individual taxpayers, to devise treatment strategies and (pre-)select taxpayers for further audits.
L. J. Kučić reports that the machine-learning model uses 158 features, derived from taxpayer historical data. New risk-indicators are immediately transmitted to the system and the compliance risk register.
Incidental reports attest that the Slovenian tax administration is experimenting with nudging tools, that adapt the language of standard communication to taxpayers to nudge them towards compliance.
What data can be processed by these systems?
The Slovenian tax administration has not disclosed the data processed by the system, during training or use.
Are these systems regulated by specific norms?
The use of machine-learning algorithm by the Slovenian tax administration is not regulated by specific legal norms.
