Metraipolis › archive › dc_67d5f65eb80e
AI Evaluation Should Work With Humans — arXiv:2608.13577
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.13577
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-19T02:20:15.081Z
- SHA-256 of the archived text
72c067be88e866db291abfab1f30700ba25ff45a113e202f7c52c7315dc91527- Stable URL of this page
https://metraipolis.com/documento/dc_67d5f65eb80e
How to cite
AI Evaluation Should Work With Humans — arXiv:2608.13577 — arXiv (en), https://arxiv.org/abs/2608.13577 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_67d5f65eb80e (Metraipolis archive dc_67d5f65eb80e, sha256 72c067be88e866db291abfab1f30700ba25ff45a113e202f7c52c7315dc91527)
BibTeX
@misc{dc67d5f65eb80e,
title = {AI Evaluation Should Work With Humans — arXiv:2608.13577},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.13577},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_67d5f65eb80e; sha256 72c067be88e866db291abfab1f30700ba25ff45a113e202f7c52c7315dc91527},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.13577. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
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. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-17 in cs.AI). Categorie arXiv: Artificial Intelligence (cs.AI) ; Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2608.13577 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-19T02:02:02.714Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-17?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).