Verifiable abstention makes AI leak diagnosis accountable in water distribution networks

Abstract:Utilities lose a substantial share of treated water to leakage, yet rarely trust artificial-intelligence localizers to dispatch crews: guessing everywhere cannot justify excavation. The gap is accountability, not accuracy: no method proves when it should not act. Here we recast leak localization as decision-making under verifiable abstention. A physics-grounded executor agent falsifies hypotheses (leak, demand, sensor, valve) against a digital twin; an independent supervisor agent, with a large-language-model (LLM) auditor, checks evidence against a code-verifiable contract, then certifies a dispatch, requests evidence or abstains. Under field-grade noise, a 32% forced baseline becomes 96% decision precision on acted events. On an independently generated benchmark it acts on only 4 of 33 leaks, all correct. A 194-event register of audited real leak locations with twin-simulated pressures and flows yields five excavation dispatches, three correct, and 44% survey recovery at full district precision. Accountable abstention offers a defensible route to autonomous water-infrastructure operation.

42 pages, 5 main figures, 1 main table, 2 extended data figures, 3 supplementary figures, 15 supplementary tables. Code and data availability described in the paper

Artificial Intelligence (cs.AI)

arXiv:2608.18836 [cs.AI]

(or arXiv:2608.18836v1 [cs.AI] for this version)

https://doi.org/10.48550/arXiv.2608.18836 arXiv-issued DOI via DataCite (pending registration)