01政策相关
5OpenAI 红队测试发现 Brotli 未认证远程代码执行漏洞并已披露
6 天前OpenAI 公布其在红队演练中发现 Brotli 压缩库存在未认证远程代码执行漏洞，攻击者可通过发送或回复恶意压缩数据在受影响系统上执行代码，无需创建用户账号，影响 Node.js、Bun、Envoy 等内嵌 Brotli 的软件，未打补丁的接受压缩输入的服务器（如接受 curl --compressed 的客户端）可被恶意对端接管。
OpenAI 详解前沿模型训练环境的安全加固与红队测试
7 天前OpenAI 介绍其前沿模型训练与评估环境的安全加固措施，包括训练期间不接入互联网、构建离线缓存网页世界、以纵深防御多层隔离，并用最强模型（含即将推出的 GPT‑6 Astra Cyber）组成智能体团队进行高算力红队测试。
OpenAI 下调估值目标至 1.25 万亿美元，拟融资至多 300 亿美元
7 天前OpenAI 接近完成至多 300 亿美元的新一轮融资，但将估值目标从最初的 1.4 万亿美元下调至 1.25 万亿美元。
国务院办公厅印发意见：增强 AI 创造就业正面效应，减缓就业替代影响
7 天前国务院办公厅 10 月 23 日印发《关于适应人工智能技术发展助力构建就业友好型发展方式的意见》，提出增强 AI 创造就业机会、赋能传统岗位的正面效应，有效减缓 AI 应用的就业替代影响。
WSJ：OpenAI 因 Signal 阅后即焚聊天解雇 3 名安全研究员
7 天前据 WSJ 报道，OpenAI 安全人员发现 3 名安全研究员与一名外部安全研究员使用阅后即焚的 Signal 聊天后将其解雇。
02行业前沿动态
4月之暗面 Kimi K3.1 Pro 上线 OpenRouter，1M 上下文多模态开源权重模型
6 天前OpenRouter 宣布月之暗面的 Kimi K3.1 Pro 上线，定位为基于 Kimi K3 后训练、面向长程编码和智能体任务的 1M 上下文多模态开源权重模型。定价为 $2/M input tokens、$10/M output tokens，可在 OpenRouter 直接试用。
月之暗面发布 Kimi K3.1 Pro，100 万 token 上下文，开放权重将于 11 月 4 日前释出
7 天前月之暗面发布 Kimi K3.1 系列首个模型 Kimi K3.1 Pro，为 3T 级开放权重模型，上下文窗口 100 万 token，并同步推出速度快 2-3 倍的 Kimi K3.1 Pro Highspeed。
Kling 4.0 在 fal 上线并向所有客户开放 API 访问
7 天前fal 宣布 Kling 4.0 正式向公众开放，所有客户均可通过 API 访问。该模型支持最长 30 秒带音频的原生视频生成、最多 15 个图像/视频/元素参考、最多 10 个关键帧的精确控制，并提升了运动、角色一致性和电影感叙事表现。
MiniMax 开放 M3.1 Flash 权重：428B 总参数、每 token 激活 23B
7 天前MiniMax 宣布 M3.1 Flash 权重公开，总参数 428B、每 token 激活 23B，提供 Hugging Face 和 ModelScope 下载。
04论文研究
9Measuring undergraduate students' reliance on generative AI during problem-solving: Scale development and validation
7 天前arXiv:2610.30195v1 Announce Type: new Abstract: Reliance on AI describes the behavioral patterns of when and how individuals depend on AI suggestions, and appropriate reliance patterns are necessary to achieve effective …
Exploring spatial configuration and group familiarity when co-watching 360-degree stories in social virtual reality
7 天前arXiv:2610.30353v1 Announce Type: new Abstract: While traditional cinematic viewing is often characterized as a social experience, engaging with immersive stories is predominantly researched as an individual activity. Th…
Atlas: Making Sense of Long Horizon Agent Trajectories
7 天前arXiv:2610.30548v1 Announce Type: new Abstract: AI agents can now search literature, generate hypotheses, analyze datasets, and run experiments at a scale difficult for researchers to match. Yet scaling scientific execut…
Beyond Better Ideas: How Human-AI Collaboration Shapes Scientific Ideas and Human Perceptions
7 天前arXiv:2610.30568v1 Announce Type: new Abstract: As AI increasingly becomes a collaborator in scientific research, its impact may extend beyond improving ideas to reshaping how they are evaluated, attributed, and perceive…
Impacts of generative AI skill development on academic adaptation and learning outcomes among university students
7 天前arXiv:2610.30599v1 Announce Type: new Abstract: Understanding how university students perceive the development of their generative artificial intelligence (GenAI) skills over time can clarify how GenAI use relates to lea…
Big World, Little Me: Comparing a First-Person VR Walkthrough Video and Static Slides for Communicating Height-Related A…
7 天前arXiv:2610.30731v1 Announce Type: new Abstract: People with dwarfism and other forms of short stature may encounter barriers created by fixed-height environments and discriminatory social assumptions. This paper evaluate…
Do We Need to "Be There" to Say Goodbye? Comparing Participatory Agency and Spatial Presence Across Digital Farewell For…
7 天前arXiv:2610.30734v1 Announce Type: new Abstract: Digitally mediated attachment is increasingly common, yet the mechanisms of symbolic closure in hybrid physical--digital relationships remain underexplored. We address an X…
Abstract Priorities, Concrete Rewards: How Stated Preferences Relate to What People Reward in AI Responses
7 天前arXiv:2610.30763v1 Announce Type: new Abstract: AI assistants built on large language models (LLMs) are trained on human feedback and increasingly personalized to users' stated preferences. Yet how closely these preferen…
Anthropic 介绍 Claude Duet 智能体的安全、隐私与凭证隔离设计
7 天前Anthropic 发文介绍处于研究预览阶段的 Claude Duet 智能体的安全与隐私设计。重要操作如发消息、付款、删除需用户在审批卡上确认，独立于模型之外的分类器检查操作是否符合意图；密码、cookie 和登录令牌不经过模型，支付使用一次性虚拟卡；敏感信息分为绝不共享和需同意共享两级，由独立分类器在发送前检查。