The rapid expansion of artificial intelligence (AI), big data (BD) and information and communication technologies (ICT) is driving substantial growth in electricity, water and material demands. Projections indicate that data centers will consume $6.7\text{--}12\%$ of U.S. electricity by 2028 (up from $4.4\%$ in 2023), while AI‑related water demand is estimated at $4\text{--}6$ billion cubic meters per year by 2027. Existing "green AI" and "red AI" debates focus mainly on computational efficiency and carbon emissions, offering limited guidance on lifecycle trade‑offs, service‑vs‑use impacts, and broader sustainability externalities beyond energy.\ \ To bridge this gap, we propose generalized green AI (GGAI) and generalized red AI (GRAI), extending the taxonomy to red BD (RBD), red ICT (RICT), red electronic and electrical engineering (REEE) and red digital technologies (RDT) within a unified Environmental‑Economic‑Social‑Technical/Operational (E/Ec/S/T) framework. Using representative footprint indicators of the large‑model era—electricity, carbon, water, and e‑waste—we analyze how risks propagate across the AI‑BD‑ICT stack and across service‑use boundaries.\ \ Actionable countermeasures are organized into a four‑tier taxonomy aligned with the E/Ec/S/T framework: (1) metric‑level reform for accountability; (2) revision of ICT design principles and demand shaping; (3) efficiency‑oriented LLM development practices covering data, model, training and scaling; and (4) energy‑infrastructure transition. The proposed framework operationalizes assessment and intervention prioritization across the AI‑BD‑ICT stacks, enabling sustainability‑by‑design without sacrificing utility.\ \ Review