Advances in high throughput and high content (HT/HC) methods such as those used in the fields of toxicogenomics, bioinformatics, and computational toxicology have the potential to improve both the efficiency and effectiveness of toxicity evaluations and risk assessments. However, prior to use, scientific confidence in these methods should be formally established. Traditional validation approaches that define relevance, reliability, sensitivity and specificity may not be readily applicable. HT/HC methods are not exact replacements for in vivo testing, and although run individually, these assays are likely to be used as a group or battery for decision making and use robotics, which may be unique in each laboratory setting. Building on the frameworks developed in the 2010 Institute of Medicine Report on Biomarkers and the OECD 2007 Report on (Q)SAR Validation, we present constructs that can be adapted to address the validation challenges of HT/HC methods. These are flexible, transparent, and require explicit specification of context and purpose of use such that scientific confidence (validation) can be defined to meet different regulatory applications. Using these constructs, we discuss how anchoring the assays and their prediction models to Adverse Outcome Pathways (AOPs) could facilitate the interpretation of results and support scientifically defensible fit-for-purpose applications.