Search papers, labs, and topics across Lattice.
CivicShield, a defense-in-depth framework inspired by diverse fields, was developed to secure government-facing AI chatbots against multi-turn adversarial attacks. It employs seven layers, including zero-trust access control, semantic firewalls, and anomaly detection, to mitigate a formal threat model encompassing eight attack families. Simulation results across 1,436 scenarios demonstrate a 72.9% detection rate with a 2.9% false positive rate, significantly outperforming single-layer defenses and effectively neutralizing multi-turn crescendo and slow-drift attacks.
Multi-layered defenses can reduce chatbot attack success rates by up to two orders of magnitude, but performance varies wildly across different benchmark suites, highlighting the need for rigorous, independent evaluation.
LLM-based chatbots in government services face critical security gaps. Multi-turn adversarial attacks achieve over 90% success against current defenses, and single-layer guardrails are bypassed with similar rates. We present CivicShield, a cross-domain defense-in-depth framework for government-facing AI chatbots. Drawing on network security, formal verification, biological immune systems, aviation safety, and zero-trust cryptography, CivicShield introduces seven defense layers: (1) zero-trust foundation with capability-based access control, (2) perimeter input validation, (3) semantic firewall with intent classification, (4) conversation state machine with safety invariants, (5) behavioral anomaly detection, (6) multi-model consensus verification, and (7) graduated human-in-the-loop escalation. We present a formal threat model covering 8 multi-turn attack families, map the framework to NIST SP 800-53 controls across 14 families, and evaluate using ablation analysis. Theoretical analysis shows layered defenses reduce attack probability by 1-2 orders of magnitude versus single-layer approaches. Simulation against 1,436 scenarios including HarmBench (416), JailbreakBench (200), and XSTest (450) achieves 72.9% combined detection [69.5-76.0% CI] with 2.9% effective false positive rate after graduated response, while maintaining 100% detection of multi-turn crescendo and slow-drift attacks. The honest drop on real benchmarks versus author-generated scenarios (71.2% vs 76.7% on HarmBench, 47.0% vs 70.0% on JailbreakBench) validates independent evaluation importance. CivicShield addresses an open gap at the intersection of AI safety, government compliance, and practical deployment.
