按年被提到
20252026 · 2 次2026
AI 安全、伦理与社会影响›通用智能与意识·arXiv:2512.10100 · cs.AI

Robust AI Security and Alignment: A Sisyphean Endeavor?

Apostol Vassilev · 2025
开始阅读 ↗PDFalphaXiv 讨论
被提及 · 2 人首次 2026-06-17 · 最近 2026-08-08发表于 IEEE Security & Privacy
2不同的人
2活跃月份
2讨论串
1被引(OpenAlex)
arXiv 摘要(候选层不写导读)

This manuscript establishes information-theoretic limitations for robustness of AI security and alignment by extending Gödel's incompleteness theorem to AI. Knowing these limitations and preparing for the challenges they bring is critically important for the responsible adoption of the AI technology. Practical approaches to dealing with these challenges are provided as well. Broader implications for cognitive reasoning limitations of AI systems are also proven.

候选层:至少 2 个不同的人、在 2 个不同月份提到过,还没到主书架的 3 人 3 个月。没有做立场标注,所以只写“被提及”。

HN 书评

推荐在前,同类按时间 · 每篇最多保留 20 条,统计用全部提及
评论sybercecurity · 2026-06

“It likely isn't: https://arxiv.org/abs/2512.10100 Feels like the days when SQL injection was seen all the time. We learned mixing instructions and data is a bad idea and a constant game of whack-a-mole. Now a new generation gets to learn this lesson.”

The White House Wants Anthropic to Block All Jailbreaks. It May Not Be Possible原评论 ↗
被提交为 story · 1 次