Scholarlink

<p>Here we quantify the material footprint of artificial intelligence training by linking computational workloads to physical hardware needs. Graphics processing unit demand is expressed as equivalent hardware lifetime consumed and, thus, the material demand involved. As computational demands rise, assessing the environmental impact of artificial intelligence requires moving beyond energy and water consumption to include the material demands of specialized hardware. We analyzed the elemental composition of a widely used graphics processing unit identifying 32 elements. The results show that artificial intelligence hardware consists of approximately 90% heavy metals and only trace amounts of precious metals. Integrating these measurements with computational throughput across varying hardware lifespans, training a large language model requires between 1760 and 8800 graphics processing units. Our findings highlight that incremental model performance gains come at disproportionately high material costs, underscoring the need to incorporate material resource considerations into discussions of artificial intelligence scalability and sustainability.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

From computation to environmental cost the resource burden of artificial intelligence

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年5月7日

查看全文
  • Sophia Falk,
  • Nicholas Kluge Corrêa,
  • Sasha Luccioni,
  • Lisa Biber-Freudenberger,
  • Aimee van Wynsberghe

摘要

Here we quantify the material footprint of artificial intelligence training by linking computational workloads to physical hardware needs. Graphics processing unit demand is expressed as equivalent hardware lifetime consumed and, thus, the material demand involved. As computational demands rise, assessing the environmental impact of artificial intelligence requires moving beyond energy and water consumption to include the material demands of specialized hardware. We analyzed the elemental composition of a widely used graphics processing unit identifying 32 elements. The results show that artificial intelligence hardware consists of approximately 90% heavy metals and only trace amounts of precious metals. Integrating these measurements with computational throughput across varying hardware lifespans, training a large language model requires between 1760 and 8800 graphics processing units. Our findings highlight that incremental model performance gains come at disproportionately high material costs, underscoring the need to incorporate material resource considerations into discussions of artificial intelligence scalability and sustainability.