人工智能作为奇异智能：反对智能的线性模型

The critique of the linear model of AI progress is endorsed and expanded upon, introducing two novel concepts: "familiar intelligence" and "strange intelligence." AI intelligence is likely to be strange intelligence, defying familiar patterns of ability and inability. It combines superhuman capacities in some domains with subhuman performance in others, and even within domains, sometimes merges superhuman insight with surprising errors that few humans would make. A nonlinear model of intelligence is developed and defended, positing that "general intelligence" is not a unified capacity but rather the ability to achieve a broad range of goals in a broad range of environments, in a manner that defies nonarbitrary reduction to a single dimension. This perspective highlights the complex and multifaceted nature of AI intelligence, moving beyond the limitations of traditional intelligence assessments. The concept of strange intelligence offers a new lens for understanding the capability boundaries of future AI systems, urging a re-evaluation of AI development trajectories. It also lays the groundwork for designing more effective assessment frameworks for artificial intelligence, acknowledging its unique and often counter-intuitive manifestations of cognitive prowess.