Algorithm Performance in Predicting Recidivism Compared to Human Judgement

A recent study by the National Institute of Justice challenges the traditional notion that AI-driven algorithms are more accurate than human judgement in predicting recidivism rates. The study found no statistically significant difference between the predictions made by the algorithm and those made by a group of people who randomly guessed recidivism outcomes. This raises questions about responsible AI governance, particularly in areas where AI is used to make critical decisions with potential implications for individuals' freedom and safety.

Matched TAIM controls

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

MEASURE 1.3 — similarity 0.600, rank 1. TAIM detail and related incidents →
MEASURE 2.6 — similarity 0.595, rank 2. TAIM detail and related incidents →
MEASURE 3.2 — similarity 0.593, 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.