Search

Materials informatics has achieved rapid progress in predicting composition–structure–property relationships, enabling accelerated screening, surrogate modeling, and exploration of high-dimensional design spaces. However, much of this success remains structurally grounded in correlational learning rather than in explanatory, transportable, or intervention-relevant forms of understanding. This conceptual manuscript argues that correlation-centric models, while often sufficient for ranking candidates under training-like conditions, are epistemically underpowered for high-stakes materials decisions such as processing optimization, microstructural control, deployment certification, and failure-sensitive design, where actions must remain defensible under distribution shift, partial observability, and changing constraints. In such settings, predictive accuracy alone does not establish decision legitimacy: a model may be correct for reasons that do not remain stable under deliberate intervention, confounding, or selection effects, thereby producing actionable recommendations without causal warrant. Motivated by recent developments in structural causal models, causal discovery, counterfactual inference, and invariant representation learning, this paper advances a theory-first reframing: materials AI should be treated as an epistemic instrument whose outputs must be qualified by the causal status they can legitimately support. We propose a novel framework—the Causal Warrant Ladder (CWL)—that classifies materials-model outputs into five ascending levels of causal legitimacy: associative regularities, transportable relations, mechanistic constraints, interventional guidance, and counterfactual design claims. CWL is paired with a Causal-Readiness Map, which specifies the minimal conceptual conditions required for upward movement on the ladder, including identifiability assumptions, invariance structure, intervention semantics, and decision stakes. By separating predictive competence from causal legitimacy, this roadmap provides a disciplined conceptual pathway beyond “black-box correlation” toward materials reasoning that supports robust and responsible design action.

This review systematically examines the conceptual treatment of causality within materials informatics literature published between 2017 and 2024, drawing exclusively on a curated set of 26 studies identified through targeted and broadened searches across Web of Science, Scopus, arXiv, and specialized databases using terms such as “causal inference,” “causality materials informatics,” “structural causal model,” “directed acyclic graph,” “intervention materials design,” and “counterfactual materials prediction,” with inclusion criteria focused on relevance to materials AI while allowing broader engineering and general causal frameworks where they intersect with materials problems. The analysis reveals a pronounced dominance of correlation-based approaches in materials artificial intelligence, where predictive models achieve impressive statistical fits for structure-property relationships yet seldom progress to robust causal claims, as evidenced by the majority of surveyed works prioritizing accuracy metrics over interventional or counterfactual reasoning. Key causal concepts and frameworks, primarily drawn from Pearl’s foundational hierarchy of association, intervention, and counterfactuals as well as structural causal models and directed acyclic graphs, are introduced and contrasted with their limited adoption in the field. Causal methods that have been applied, albeit sparingly, to materials informatics—ranging from data-driven causal discovery to Bayesian causal modeling—are surveyed alongside their strengths and context-specific limitations. Persistent challenges, including the rarity of randomized interventions in experimental materials workflows and the confounding effects inherent in high-dimensional observational datasets, are highlighted as barriers that leave substantial gaps in the literature. Ultimately, this review offers targeted recommendations for authors, reviewers, and the broader community to integrate causal reasoning more explicitly, thereby moving materials informatics from correlational prediction toward actionable intervention and counterfactual understanding essential for autonomous materials design.