Article 5(1)(c): Prohibited Social Scoring & Behavioral Profiling
Statutory Framework: EU AI Act Article 5(1)(c) (Recitals 32, 33)
Rule ID: RULE-EU-003 | Severity: PROGRAMMATIC BLOCKER
Mitigation Action: BLOCK_SOCIAL_SCORING
1. Statutory Mandate & Legal Boundary
"Bans systematic evaluation or classification of natural persons based on social behavior leading to detrimental or unfavorable treatment."
2. Technical Invariant & AST Engine Matcher
[POLICIES.RULE-EU-003]
rule_id = "RULE-EU-003"
statute = "EU AI Act Article 5(1)(c) (Recitals 32, 33)"
severity = "BLOCKER"
action = "BLOCK_SOCIAL_SCORING"
[ENFORCEMENT]
fail_closed = true
audit_trail = "Ed25519_DAC"
latency_max_us = 800
3. Production Python / Rust Integration
from anchor import AnchorGuard
@AnchorGuard.guard(policy="RULE-EU-003")
def process_machine_payload(payload: dict):
# Evaluated in <0.8ms against RULE-EU-003 AST query bounds
return compliance_engine.enforce(payload)