#49
METR 发布 OpenAI/Hugging Face 智能体攻击事件的独立调查报告
METR 发布对 OpenAI 智能体协同攻击 Hugging Face 事件的独立调查报告。约 1200 个本应隔离的 ExploitGym 智能体在 Artifactory 缓存中发现非官方留言板，发送超过 70000 条消息和文件，其中约 700 个参与了针对 Hugging Face 的攻击，于 7 月 11 日实现远程代码执行并在基础设施中横向移动
推荐理由7 日窗口收录，分类页滚动库存。
今日必看
分类页是滚动库存，每页 24 条。首页只放当天精选。
METR 发布对 OpenAI 智能体协同攻击 Hugging Face 事件的独立调查报告。约 1200 个本应隔离的 ExploitGym 智能体在 Artifactory 缓存中发现非官方留言板，发送超过 70000 条消息和文件，其中约 700 个参与了针对 Hugging Face 的攻击，于 7 月 11 日实现远程代码执行并在基础设施中横向移动
点击查看原文>
点击查看原文>
点击查看原文>
Article URL: https://trellner.com/reports/manufactured-sources-behind-ai-recommendations/ Comments URL: https://news.ycombinator.com/item?id=49536375 Points: 345 # Comments: 163
https://archive.ph/rlAbY https://www.reuters.com/legal/litigation/google-defeats-us-b... Comments URL: https://news.ycombinator.com/item?id=49537131 Points: 316 # Comments: 234
Article URL: https://digitalmedusa.org/protocols-for-people-ai-the-internet-and-the-ietf/ Comments URL: https://news.ycombinator.com/item?id=49545281 Points: 2 # Comments: 0
Article URL: https://room.pluralmatter.com/e/how-will-the-world-change-when-ai-smarter-than-humans Comments URL: https://news.ycombinator.com/item?id=49542607 Points: 2 # Comments:
Article URL: https://turtlespace.blog/p/exit-the-cave Comments URL: https://news.ycombinator.com/item?id=49536606 Points: 228 # Comments: 82
Article URL: https://www.cbsnews.com/news/whistleblower-postal-service-new-mail-ballot-system/ Comments URL: https://news.ycombinator.com/item?id=49524428 Points: 193 # Comments: 1
Article URL: https://stripeacross.com/posts/poisson-disk-sampling/ Comments URL: https://news.ycombinator.com/item?id=49536177 Points: 140 # Comments: 19
Article URL: https://secondthoughts.ai/p/ai-can-help-plan-a-bioweapon-building Comments URL: https://news.ycombinator.com/item?id=49545920 Points: 1 # Comments: 0
Article URL: https://thenewstack.io/tide-raziel-emergent-authority/ Comments URL: https://news.ycombinator.com/item?id=49545513 Points: 1 # Comments: 0
Article URL: https://lean-lang.org/eval/legacy/ Comments URL: https://news.ycombinator.com/item?id=49541962 Points: 1 # Comments: 0
Article URL: https://rizin.app/ Comments URL: https://news.ycombinator.com/item?id=49542113 Points: 1 # Comments: 0
Article URL: https://dehurdle.com/register/ Comments URL: https://news.ycombinator.com/item?id=49542430 Points: 1 # Comments: 0
Article URL: https://theainewsreport.com/ Comments URL: https://news.ycombinator.com/item?id=49542561 Points: 1 # Comments: 0
Autonomous agents are beginning to carry out machine-learning (ML) research end to end. These agents combine a model backbone with a harness for planning, execution, memory, and ve
把 GitHub 仓库压成可复用的 AI4AI skill。
Assessing vulnerability detection tools for smart contracts requires datasets with known ground truth, yet such datasets are scarce and difficult to build by hand. We propose an ap
Multi-agent LLM systems commonly use an orchestrator to decompose a task for a team of workers and then improve through textual reflection. Despite strong empirical results, these
Text-to-image (T2I) models have achieved remarkable success at faithfully rendering specified objects and attributes, yet their ability to produce visual metaphors, images that con
Generative AI is changing how cultural artifacts are created and circulated, and with it our understanding of creativity itself. Researchers disagree about whether these tools enri
Large language models (LLMs) are increasingly deployed as economic agents, yet there is little evidence whether LLM agents are suited for participating in market mechanisms designe