跳到正文
Patterns· Johnson·收录于· 6 天前

Competition for attention predicts good-to-bad tipping in AI

Competition for attention predicts good-to-bad tipping in AI

原文摘要

出版方登记 · OpenAlex

More than 50% of the world's population carry devices that can run personal AI companions with no Internet connection and hence limited safety checks. Here, we discuss how a large class of potentially dangerous good-to-bad output tipping in such edge AI can arise at the atomistic scale (i.e., attention head) through competition for the machinery's attention between the conversation's context and competing output basins. We derive an approximate closed-form estimate for the dynamical tipping point denoted as n ∗ here. We present evidence consistent with the predicted immediate versus delayed/no-tipping regime dynamics across seven 124 million–12 billion parameter open-weight models as well as from conversations in independent production-scale studies.

页面未能读取全文,以上为出版方登记的摘要;方法与图表见原文

来源:Patterns · doi.org