INTRODUCTION: Insulin autoantibodies (IAA) are key predictors of type 1 diabetes, particularly in young children. Micro-radiobinding assays (RBA) are the gold standard for IAA measurement but have limitations. We assessed whether a luciferase immunoprecipitation system (LIPS) assay improved diabetes risk assessment. METHODS: To validate LIPS compared with RBA, samples from people with new-onset type 1 diabetes (n = 150) and first-degree relatives (FDRs) (n = 619), of whom 91 had developed diabetes during follow-up, were used. This cross-sectional observational data was analysed using the area under the receiver operator characteristic curve and cox-proportional hazard models. RESULTS: In new-onset diabetes, RBA and LIPS showed 88% agreement in IAA status. Positive IAA LIPS was more common in 89 FDRs with high-moderate affinity IAA (61%) compared with 22 FDRs with low-affinity IAA (18%) (p < 0.001). In FDRs positive for multiple other islet autoantibodies, 20-year diabetes risk was 80% for those positive compared with 30% for those negative for IAA by LIPS (p = 0.013). IAA LIPS added to diabetes risk independently of status/level of IAA by RBA, other autoantibodies and sampling age (p < 0.001). CONCLUSION: The IAA LIPS low-blood-volume, high-throughput technique identifies more individuals with the highest risk of diabetes. The ability to identify high-affinity IAA makes LIPS an ideal method for future clinical trials and population screening strategies to predict the risk of diabetes.