Engineering synthetic biology sensors with artificial intelligence: From programmable circuits to next-generation biosensing.

Artificial intelligence (AI) is advancing synthetic biology biosensors (SBBs), driving a fundamental shift from rational design to AI-driven prediction. This review establishes a systematic framework linking AI algorithms to the Design-Build-Test-Learn (DBTL) cycle. We explicitly analyze the engineering paradigms of AI-enabled cell-based SBBs and AI-optimized cell-free SBBs, highlighting how computational intelligence addresses platform-specific bottlenecks. Crucially, we synthesize the AI-driven workflow into three core frontiers: AI-guided robust sensor element design, AI-assisted signal processing for accurate performance characterization, and AI-driven closed-loop optimization to accelerate autonomous evolution. Furthermore, representative applications of SBBs are investigated, including multi-pollutant environmental detection, continuous biomarker monitoring, food safety tracking, and intelligent biomanufacturing. Beyond achievements, we critically evaluate unresolved obstacles, notably the "reality gap" and the "small-data dilemma". Finally, we propose a roadmap centered on bio-digital hybrid interfaces, explainable AI, and data standardization to accelerate the transformation of SBBs into robust, field-deployable intelligent sensing systems.