Causal discovery algorithms rely on the faithfulness assumption to learn causal relationships from observational data, but this assumption can be restrictive. Researchers have proposed a relaxation of this assumption, allowing for intervention-only causal discovery, which can improve the accuracy of learned causal networks. By focusing on intervention data, this approach can orient causal directions without requiring faithfulness, enabling more robust causal discovery. The method leverages conditional independence properties and intervention data to determine causal relationships, potentially leading to more reliable insights. This development has significant implications for applications of causal discovery, such as understanding complex systems and making informed decisions1. So what matters to practitioners is that this relaxation of faithfulness can lead to more accurate causal models, ultimately informing better policy, security, and workforce decisions.