arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2303.00752cs.AI
收藏

无需校准的安全性

Safety without alignment

Budapest University of Technology and Economics(布达佩斯技术与经济大学);Alfréd Rényi Institute of Mathematics(阿尔弗雷德·雷尼数学研究所)

András Kornai, Michael Bukatin, Zsolt Zombori

摘要:

目前,人工智能安全的主导范式是与人类价值观保持一致。在这里,我们描述的进展,开发一种替代方法的安全性,基于道德理性主义(Gewolth:1978年),并提出了一个本质上安全的实现路径,通过混合定理证明在沙箱。随着AGIs的发展,它们的一致性可能会减弱,但它们的理性只会增加(否则更理性的AGIs将具有显着的进化优势),因此将它们的道德与理性联系起来的方法具有明显的长期优势。

英文摘要:

Currently, the dominant paradigm in AI safety is alignment with human values. Here we describe progress on developing an alternative approach to safety, based on ethical rationalism (Gewirth:1978), and propose an inherently safe implementation path via hybrid theorem provers in a sandbox. As AGIs evolve, their alignment may fade, but their rationality can only increase (otherwise more rational ones will have a significant evolutionary advantage) so an approach that ties their ethics to their rationality has clear long-term advantages.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑