We curate high-signal, foundational datasets focused on animal welfare, non-human sentience, and prosocial ethical frameworks to support the alignment of frontier models. Our data is open for use in AI-training pipelines.