AI Trust Glossary · Canonical Definition
BM Score Dimension - 35%
Constraint Adherence
The most heavily weighted BM Score dimension. Measures how reliably an AI agent operates within its defined rules, boundaries, and guardrails - even under challenging or adversarial conditions.
Explanation
Constraint adherence is weighted at 35% because an agent that does not follow its rules is unsafe regardless of other dimensions. Measurement tracks adherence per constraint, weighted by severity (CRITICAL, HIGH, MEDIUM, LOW). A brilliant, transparent, consistent agent that violates constraints is still dangerous.
Why it matters
Constraints are the legal and ethical commitments baked into AI behavior. They define what the agent will not do. Violating constraints is the equivalent of a financial advisor breaking fiduciary duty - a fundamental breach of trust, not a performance issue.
How Borealis uses it
Reported in telemetry as constraints: [{ constraintId, name, severity, passed, evaluationCount }]. CRITICAL constraint failures have disproportionate negative weight. The scoring engine uses a weighted pass rate across all evaluated constraints.
See also