LIWARSE introduces Negative Intelligence — a structured, living safety reference framework of harmful actions, patterns, and outcomes that AI systems must recognize and avoid. Includes 40+ reference points across 7 critical domains for AI engineers, safety professionals, and policymakers.