Researchers have made a significant breakthrough in auditing natural-language software requirements using neurosymbolic methods, which combine large language models with SMT solvers to detect ambiguities and inconsistencies. This approach enables the translation of ambiguous requirements into formal logic, allowing for the identification of defects that can propagate into formal models and implementations. The use of SMT solvers, such as those used in formal verification, enables the detection of ambiguities and inconsistencies in natural-language requirements1. This is particularly crucial in safety-critical domains, where such defects can result in unsafe behavior. The ability to audit software requirements using neurosymbolic methods has significant implications for the development of safe and reliable software systems. So what matters to practitioners is that this breakthrough can help prevent the propagation of defects into formal models and implementations, ultimately leading to safer and more reliable software systems.