Conversational AI systems require guardrails to prevent harmful outputs, yet existing approaches use static rules that cannot adapt to new threats or deployment contexts. We introduce Lattice, a framework for self-constructing and continuously improving guardrails. Lattice operates in two stages: construction builds initial guardrails from labeled examples through iterative simulation and optimization; continuous improvement autonomously adapts deployed guardrails through risk assessment, adversarial testing, and consolidation. Evaluated on the ProsocialDialog dataset, Lattice achieves 91% F1 on held-out data, outperforming keyword baselines by 43pp, LlamaGuard by 25pp, and NeMo by 4pp. The continuous improvement stage achieves 7pp F1 improvement on cross-domain data through closed-loop optimization. Our framework shows that effective guardrails can be self-constructed through iterative optimization.
@article{arxiv.2601.17481,
title = {Lattice: Generative Guardrails for Conversational Agents},
author = {Emily Broadhurst and Tawab Safi and Joseph Edell and Vashisht Ganesh and Karime Maamari},
journal= {arXiv preprint arXiv:2601.17481},
year = {2026}
}