Two methods for classifying Adult Attachment Interviews (AAI) for family court decisionmaking, Berkeley and DMM, differ in their usefulness for family court. Both expand Ainsworth's three ABC infant categories. The Berkeley method adds a fourth Disorganised/Unresolved category in adulthood, but has low specificity and validity for risk parents. The DMM method identifies an expanding array of strategies across the lifespan, as neurological development makes more complex strategies possible. This study examined DMM-AAI classifications in a sample of 332 British AAIs and compared the results to published meta-analyses of the Berkeley AAI. Six a priori hypotheses addressed the central question raised: which classificatory method for the AAI is more useful for child protection? DMM-AAI classifications differentiated between (i) normative adults, (ii) parents with mental health problems, (iii) parents in family court proceedings and (iv) incarcerated violent criminals on attachment, psychological trauma and pervasively high or low arousal. We assert that the DMM-AAI is sufficiently valid and discriminating for court use and that it can contribute to court decision-making when integrated with other assessments and clinical reports.