Position Paper Advocates for Planet-Centered AI Design Philosophy

This position paper introduces Planet-Centered AI (PCAI) as a new design philosophy and research agenda, arguing that current AI paradigms, which are largely human-centered, are insufficient for tackling complex global issues. PCAI advocates for reorienting AI development to focus on planetary-scale socio-ecological systems and their long-term dynamics. The paper diagnoses limitations in existing AI frameworks, particularly their human-centric biases, which become problematic when dealing with systemic risks, non-stationarity, and deep uncertainty characteristic of current planetary conditions. It emphasizes a systems-thinking approach, viewing Earth as an interconnected whole where humans are an integral part. PCAI proposes reshaping the entire AI lifecycle, from problem formulation and model design to evaluation and deployment. This involves aligning AI with global agendas, developing system-aware AI foundations, using trajectory-oriented evaluation, and ensuring monitorability. The authors claim that AI systems optimized without explicit consideration of systemic consequences are more likely to worsen systemic instability rather than mitigate it.