Algorithmic risk assessment tools now influence bail, sentencing, and parole for approximately two-thirds of the US population. Most of the people affected don't know these tools are used. Most institutions using them have no independent evaluation of their performance.
Risk assessment algorithms are sold as objective replacements for subjective judicial bias. The evidence does not support this framing. Dressel and Farid (2018) demonstrated that COMPAS performs no better than untrained lay prediction while producing racial disparities in false-positive rates. Berk et al. (2024) showed that the bias originates in the training data — arrest records encoding decades of racially disparate policing — making technical corrections insufficient. The governance response is to build the institutional infrastructure that makes these tools accountable.
COMPAS accuracy 65.2% vs. lay prediction 67.0%. Black defendants flagged as high-risk at 2× the false-positive rate of white defendants.
The bias-in-bias-out problem. Training data encodes racially disparate policing. Technical corrections cannot solve data-origin bias.
Formal proof: calibration and equal false-positive rates cannot simultaneously hold when base rates differ across groups.
The prototype organizes reflection across six proposed governance dimensions. It does not produce a validated institutional score, audit, accreditation, or certification.
Plain-language documentation of all AI tools, publicly available; independent demographic impact assessment specific to the jurisdiction's population; override protocol with no coercive documentation burden; notification to affected individuals; error classification and remediation process; community advisory body with authority to recommend discontinuation.
The full tier operationalization — PIONEERING through EARLY STAGE, with specific observable criteria for all six dimensions — is documented in the Framework Foundation.
Court innovation offices · Legal aid tech teams · Public defender AI tool governance · Access-to-justice organizations · Bar association AI policy committees
And anyone a risk score is run on. Most people scored by these tools never know. If a decision about your bail, sentence, or parole runs through an algorithm, this governance is your recourse.
Review the module scope and proposed dimensions. The assessment remains an educational prototype; no email report or institutional finding is active.