Scholarlink

<p>Understanding the risk from applications of artificial intelligence (AI) is a critical part of creating AI governance strategies. Building on the idea of studying AI using ecological and evolutionary perspectives, we propose a novel approach for assessing risk from AI using indicators derived from theoretical ecology models. We illustrate our methods by deriving 3 indicators from population and ecosystem models originating from theoretical ecology. We conclude with a discussion of limitations of our analysis and considerations for improving AI governance policy.</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

The ecology of AI risk

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年6月25日

查看全文
  • Edward Geist,
  • Alexander Dolnick Meyer,
  • Alvin Moon,
  • Aisha Nájera,
  • James Holland Jones,
  • Anton Wu

摘要

Understanding the risk from applications of artificial intelligence (AI) is a critical part of creating AI governance strategies. Building on the idea of studying AI using ecological and evolutionary perspectives, we propose a novel approach for assessing risk from AI using indicators derived from theoretical ecology models. We illustrate our methods by deriving 3 indicators from population and ecosystem models originating from theoretical ecology. We conclude with a discussion of limitations of our analysis and considerations for improving AI governance policy.