1
Anthropic 安全框架自动降级：Claude 误删开发者 700 GB 主目录，自动化清理反成灾难
IT之家 8 月 29 日消息，开发者 Sebastien Guillemot 本周三在 X 上反馈称，他在测试 AI 智能体的文件删除防护机制时遭遇严重事故，Claude 直接给他删除了大约 700GB 数据，包括整个用户主目录，相当于一周的工作成果。 据称，事故发生前，Anthropic 的安全…
2
AEGIS: Preventing Cross-Domain Resource Abuse in MCP
首个针对MCP跨域资源滥用防护框架登场，为AI协议安全加装“门禁系统”
arXiv:2608.20481v1 Announce Type: cross Abstract: The Model Context Protocol (MCP) is an open source JSON-RPC protocol that standardizes how large lan…
3
1Password for Claude 上线：AI 交互场景下用户无需手动填写密码
1Password与Claude集成，基于零暴露安全框架让AI自动填密码，提升交互安全与效率
IT之家 7 月 17 日消息，密码管理工具 1Password 昨日（7 月 16 日）发布公告，宣布携手 Anthropic 公司，推出 1Password for Claude ，让聊天机器人在用户授权后调用已存储的登录凭证，完成预订行程、管理在线账户等多步骤任务。 IT之家附上相关演示视频如…
4
5
6
Harness-MU: A Safe, Governed, and Effective Harness for Multi-User LLM Agents
多用户LLM Agent的安全治理框架Harness-MU，保障可控与高效。
arXiv:2606.21856v1 Announce Type: cross Abstract: The increasing deployment of large language model (LLM) agents in collaborative workflows demands ro…
7
A Layered Security Framework Against Prompt Injection in RAG-Based Chatbots
了解RAG聊天机器人防注入攻击的分层安全框架，开放获取的学术前沿。
arXiv:2606.19660v1 Announce Type: cross Abstract: Prompt injection is ranked as the most critical vulnerability in large language model (LLM) deployme…
8
Automated Framework to Evaluate and Harden LLM System Instructions against Encoding Attacks
提出自动化框架，系统评估并加固LLM系统指令以抵御编码攻击，为AI安全提供新工具。
arXiv:2604.01039v2 Announce Type: replace-cross Abstract: System Instructions in Large Language Models (LLMs) are commonly used to enforce safety poli…
9
Auditing Agent Harness Safety
这篇论文系统审计了智能体框架的安全隐患，为构建可信AI系统提供关键方法论。
arXiv:2605.14271v2 Announce Type: replace Abstract: LLM agents increasingly run inside execution harnesses that dispatch tools, allocate resources, an…
10
Operator System Card
OpenAI 发布 Operator 系统卡，详解多层级安全防护与红队测试成果。
Drawing from OpenAI’s established safety frameworks, this document highlights our multi-layered approach, including model and product mitigations we’v…
11
Working with US CAISI and UK AISI to build more secure AI systems
OpenAI与美英AI安全机构合作，共建更安全的AI系统，最新进展一览。
OpenAI shares progress on the partnership with the US CAISI and UK AISI to strengthen AI safety and security.