Canada - Criminal Justice Risk Assessment Algorithm

What Happened

In 2021, Canadian civil liberties groups revealed that correctional services in several provinces utilized automated algorithmic risk-assessment tools (such as the Service Planning Instrument - SPI) to determine recidivism risk and custody classification for indigenous prisoners. The system combined historical demographic data and socio-economic indicators to output "risk scores" that directly influenced parole decisions and security classification. Investigations proved that the algorithm systematically over-classified indigenous prisoners as "high risk," trapping them in maximum-security facilities at higher rates than non-indigenous prisoners with identical criminal histories due to systemic bias in the historical training data.

How MEG 1 would have acted

The risk-assessment tool failed the Index of Safety and Responsibility (ISR) audit due to systemic racial discrimination and low accuracy in predicting actual behavioral outcomes, violating the requirements of procedural fairness.

MEG 2 Analysis - Legal Framework

Agent level: N2 (management) - automated scoring and risk-profiling system used by a public authority to make decisions directly affecting judicial classification and freedom of movement.
Cause (6.1):
(a) system defect - algorithmic bias and systemic discrimination against indigenous populations built into the risk model (design defect), attributable to the manufacturer.
(b) autonomous decision error - the system's decision to assign inflated risk scores based on biased demographic proxies.
Attachment of liability: The correctional services as the operator are fully liable for the wrongful judicial classifications caused by deploying an unverified, biased system.

Procedural Mechanism

6.2 (architectural human confirmation): Decisions involving security classification or parole eligibility must require active, architectural human confirmation. The AI must act as a advisory tool; a human judicial board must independently verify and document the justification for the final score.
7.1 (independent forensics): The input parameters, the risk scoring weights, and the system's decision path must be securely logged in an unalterable ledger to allow for judicial and human rights audits.
9.4 (least privilege): The profiling system was granted significant influence over prisoner liberty without public validation or independent oversight layers, violating safe operational boundaries.