Developing economies are characterized by limited compliance with government regulations, such as taxation. Resources for enforcement are scarce and audit cases are often selected in a discretionary manner. We study whether the increasing availability of digitized data helps improve audit targeting. In a field experiment at scale in Senegal, we compare tax audits selected by inspectors to audits selected by a risk-scoring algorithm. We find that inspector-selected audits are more likely to be conducted and uncover more evasion. However, algorithm-selected audits require less manpower and may generate less corruption. In ongoing work, we attempt to unpack the algorithm’s (dis)functioning and its interaction with human capital.