Are Bail Algorithms as Effective as Random Guesses?

A recent study published in the Proceedings of Machine Learning Research has questioned the accuracy of bail algorithms used by courts across the US. The research found that these algorithms, designed to predict whether a defendant will appear at their trial or commit another crime while out on bail, are no more accurate than random guesses. This finding raises concerns about the fairness and accountability of AI systems in critical decision-making processes.

Matched TAIM controls

Suggested mapping from embedding similarity (not a formal assessment). Browse all TAIM controls

MEASURE 1.3 — similarity 0.587, rank 1. TAIM detail and related incidents →
MEASURE 3.2 — similarity 0.583, rank 2. TAIM detail and related incidents →
MEASURE 2.5 — similarity 0.572, rank 3. TAIM detail and related incidents →

Source

Data from the AI Incident Database (AIID). Cite this incident: https://incidentdatabase.ai/cite/40

Data source

Incident data is from the AI Incident Database (AIID).

When citing the database as a whole, please use:

McGregor, S. (2021) Preventing Repeated Real World AI Failures by Cataloging Incidents: The AI Incident Database. In Proceedings of the Thirty-Third Annual Conference on Innovative Applications of Artificial Intelligence (IAAI-21). Virtual Conference.

Pre-print on arXiv · Database snapshots & citation guide

We use weekly snapshots of the AIID for stable reference. For the official suggested citation of a specific incident, use the “Cite this incident” link on each incident page.