AI Evaluation Should Work With Humans

作者: Jan Kulveit, Gavin Leech, Tomáš Gavenčiak, Raymond Douglas

摘要

This position paper argues that the dominant paradigm of AI evaluation (which focuses on superhuman autonomous performance and so implicitly targets the goal of replacing humans) is guiding AI development in the wrong direction. Instead, the AI community should pivot to evaluating the performance of human--AI teams. We argue that this collaborative shift will foster AI systems that act as true complements to human capabilities and therefore lead to far better societal outcomes than will the current process.

打开原文arXiv
用 AI 找相似论文

可以直接复制这句话问 AI

用 Paperlist 找和「AI Evaluation Should Work With Humans」相似的论文(id: 86291,arXiv 2026)
set up https://paperlist.ai/SKILL.md

接入 MCP命令行 CLI