<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/">
<channel>
<title>AI 资讯 · OrererereO</title>
<link>https://orererereo.me/ai-news</link>
<description>官方发布与论文，整理成可读的中文摘要。</description>
<language>zh-CN</language>
<lastBuildDate>Sat, 07 Nov 2026 00:00:00 GMT</lastBuildDate>
<atom:link href="https://orererereo.me/ai-news.xml" rel="self" type="application/rss+xml"/>
<item>
<title>Gemini Omni 1.2 Flash: Fast and cost-effective with longer clips</title>
<link>https://deepmind.google/blog/gemini-omni-12-flash-fast-and-cost-effective-with-longer-clips/</link>
<guid isPermaLink="true">https://deepmind.google/blog/gemini-omni-12-flash-fast-and-cost-effective-with-longer-clips/</guid>
<pubDate>Fri, 06 Nov 2026 00:00:00 GMT</pubDate>
<category>Google DeepMind</category>
<description>Google DeepMind 发布 Gemini Omni 1.2 Flash，官方称其快速且成本效益高，并支持更长的片段。</description>
</item>
<item>
<title>Gemini Omni 1.2 Flash: Fast and cost-effective with longer clips</title>
<link>https://blog.google/innovation-and-ai/technology/developers-tools/introducing-gemini-omni-1-2-flash/</link>
<guid isPermaLink="true">https://blog.google/innovation-and-ai/technology/developers-tools/introducing-gemini-omni-1-2-flash/</guid>
<pubDate>Fri, 06 Nov 2026 00:00:00 GMT</pubDate>
<category>Google AI</category>
<description>Gemini Omni 1.2 Flash 支持原生 15 秒视频生成与编辑，一致性更强，价格降低 30%。</description>
</item>
<item>
<title>Partnering with Hacktron to find vulnerabilities in OpenAI’s systems</title>
<link>https://openai.com/index/partnering-with-hacktron</link>
<guid isPermaLink="true">https://openai.com/index/partnering-with-hacktron</guid>
<pubDate>Fri, 06 Nov 2026 00:00:00 GMT</pubDate>
<category>OpenAI</category>
<description>OpenAI 与 Hacktron 合作，寻找 OpenAI 系统中的漏洞、测试与第三方服务的连接，并利用前沿 AI 加强防御。</description>
</item>
<item>
<title>Measuring how AI agents handle everyday wet-lab science</title>
<link>https://openai.com/index/measuring-how-ai-agents-handle-wet-lab-science</link>
<guid isPermaLink="true">https://openai.com/index/measuring-how-ai-agents-handle-wet-lab-science</guid>
<pubDate>Fri, 06 Nov 2026 00:00:00 GMT</pubDate>
<category>OpenAI</category>
<description>OpenAI 推出 BioBasicsBench，这是一个包含 100 项任务的基准，用于衡量 AI agent 解读原始实验数据和处理日常湿实验室科学工作的能力。</description>
</item>
<item>
<title>Setting a new standard for responsible AI infrastructure</title>
<link>https://openai.com/index/setting-new-standard-responsible-ai-infrastructure</link>
<guid isPermaLink="true">https://openai.com/index/setting-new-standard-responsible-ai-infrastructure</guid>
<pubDate>Thu, 05 Nov 2026 00:00:00 GMT</pubDate>
<category>OpenAI</category>
<description>OpenAI 发布 National Community Compact 和 Compute Campus Policy Blueprint，称其用于保护当地资源、创造就业，并指导负责任的 AI 基础设施建设。</description>
</item>
<item>
<title>Le Chonk brings the brains. Studio puts it to work.</title>
<link>https://mistral.ai/news/build-with-studio/</link>
<guid isPermaLink="true">https://mistral.ai/news/build-with-studio/</guid>
<pubDate>Thu, 05 Nov 2026 00:00:00 GMT</pubDate>
<category>Mistral</category>
<description>Mistral 推出 Mistral Studio 和 Le Chonk，官方称可更快构建和部署企业级 agent、工作流和应用，并对模型、数据和基础设施拥有完全控制权。</description>
</item>
<item>
<title>Agentic cheminformatics with Google Antigravity</title>
<link>https://deepmind.google/blog/agentic-cheminformatics-with-google-antigravity/</link>
<guid isPermaLink="true">https://deepmind.google/blog/agentic-cheminformatics-with-google-antigravity/</guid>
<pubDate>Wed, 04 Nov 2026 00:00:00 GMT</pubDate>
<category>Google DeepMind</category>
<description>Google DeepMind 介绍 Google Antigravity 与 Cheminformatics Skills 如何加速药物发现工作流、搜索化合物并提出新型刚性化大环化合物。</description>
</item>
<item>
<title>OpenAI to acquire Pierre Computer Company, Inc.</title>
<link>https://openai.com/index/openai-to-acquire-pierre</link>
<guid isPermaLink="true">https://openai.com/index/openai-to-acquire-pierre</guid>
<pubDate>Wed, 04 Nov 2026 00:00:00 GMT</pubDate>
<category>OpenAI</category>
<description>OpenAI 将收购 Pierre Computer Company，以扩展 Codex 用于构建、理解和审查代码的工具。</description>
</item>
<item>
<title>Daybreak Red is now available in AWS GovCloud (US-East)</title>
<link>https://openai.com/index/daybreak-red-aws-govcloud</link>
<guid isPermaLink="true">https://openai.com/index/daybreak-red-aws-govcloud</guid>
<pubDate>Wed, 04 Nov 2026 00:00:00 GMT</pubDate>
<category>OpenAI</category>
<description>Daybreak Red 通过 Amazon Bedrock 在 AWS GovCloud (US-East) 向获批客户提供 GPT-5.6 Cyber，用于授权的网络安全工作。</description>
</item>
<item>
<title>A thousand fewer paths to root</title>
<link>https://openai.com/index/a-thousand-fewer-paths-to-root</link>
<guid isPermaLink="true">https://openai.com/index/a-thousand-fewer-paths-to-root</guid>
<pubDate>Wed, 04 Nov 2026 00:00:00 GMT</pubDate>
<category>OpenAI</category>
<description>OpenAI 构建了一条自主流水线，在 Linux 内核中发现、验证并生成了 948 个本地权限提升漏洞的补丁。</description>
</item>
<item>
<title>Introducing Images 2.6</title>
<link>https://openai.com/index/introducing-images-2-6</link>
<guid isPermaLink="true">https://openai.com/index/introducing-images-2-6</guid>
<pubDate>Wed, 04 Nov 2026 00:00:00 GMT</pubDate>
<category>OpenAI</category>
<description>OpenAI 发布 Images 2.6，在 ChatGPT、Codex 和 API 中带来更快的图像生成、GIF 创建、更精确的编辑以及新的 API 控制。</description>
</item>
<item>
<title>Rewarding defenders with bigger bug bounties</title>
<link>https://openai.com/index/rewarding-defenders-with-bigger-bug-bounties</link>
<guid isPermaLink="true">https://openai.com/index/rewarding-defenders-with-bigger-bug-bounties</guid>
<pubDate>Tue, 03 Nov 2026 00:00:00 GMT</pubDate>
<category>OpenAI</category>
<description>OpenAI 提高 Security Bug Bounty 奖励，并为独立安全研究人员新增报告质量、忠诚度和优先活动奖金。</description>
</item>
<item>
<title>Preparing our next security sprint to harden OpenAI</title>
<link>https://openai.com/index/preparing-our-next-security-sprint-to-harden-openai</link>
<guid isPermaLink="true">https://openai.com/index/preparing-our-next-security-sprint-to-harden-openai</guid>
<pubDate>Tue, 03 Nov 2026 00:00:00 GMT</pubDate>
<category>OpenAI</category>
<description>OpenAI 正与 Cathedral Systems 筹备下一轮内部安全冲刺，利用 AI 测试并加强自身防御。</description>
</item>
<item>
<title>OpenAI appoints Jill Hazelbaker as Chief Corporate Affairs Officer</title>
<link>https://openai.com/index/jill-hazelbaker-chief-corporate-affairs-officer</link>
<guid isPermaLink="true">https://openai.com/index/jill-hazelbaker-chief-corporate-affairs-officer</guid>
<pubDate>Tue, 03 Nov 2026 00:00:00 GMT</pubDate>
<category>OpenAI</category>
<description>Jill Hazelbaker 将加入 OpenAI 担任首席企业事务官，负责领导公司的全球传播与企业定位。</description>
</item>
<item>
<title>Introducing the Grid Protectors Program</title>
<link>https://openai.com/index/grid-protectors-program</link>
<guid isPermaLink="true">https://openai.com/index/grid-protectors-program</guid>
<pubDate>Mon, 02 Nov 2026 00:00:00 GMT</pubDate>
<category>OpenAI</category>
<description>OpenAI 推出 Grid Protectors Program，将与公用事业公司合作，帮助未来的计算园区支持电网可靠性并保护消费者电费。</description>
</item>
<item>
<title>LifeVal</title>
<link>https://openai.com/index/lifeval</link>
<guid isPermaLink="true">https://openai.com/index/lifeval</guid>
<pubDate>Mon, 02 Nov 2026 00:00:00 GMT</pubDate>
<category>OpenAI</category>
<description>OpenAI 推出 LifeVal，这是一个基准，用于衡量 AI 在帮助人们完成日常任务方面的表现，涵盖寻找托儿服务、管理财务和处理政府文书等场景。</description>
</item>
<item>
<title>Scaling Git at OpenAI</title>
<link>https://openai.com/index/scaling-git-at-openai</link>
<guid isPermaLink="true">https://openai.com/index/scaling-git-at-openai</guid>
<pubDate>Mon, 02 Nov 2026 00:00:00 GMT</pubDate>
<category>OpenAI</category>
<description>OpenAI 介绍了 Pando 的构建方式，这是一个兼容 Git 的后端，为 Codex 所支持的大规模开发而设计。</description>
</item>
<item>
<title>OpenAI appoints Dan Driscoll as Senior Counselor for National Security Modernization</title>
<link>https://openai.com/index/dan-driscoll-joins-openai</link>
<guid isPermaLink="true">https://openai.com/index/dan-driscoll-joins-openai</guid>
<pubDate>Mon, 02 Nov 2026 00:00:00 GMT</pubDate>
<category>OpenAI</category>
<description>OpenAI 任命 Dan Driscoll 为国家安全现代化高级顾问，负责就国家安全现代化、网络韧性以及与美国的盟友合作中的负责任 AI 提供建议。</description>
</item>
<item>
<title>An updated Model Spec for more capable and beneficial AI</title>
<link>https://openai.com/index/updated-model-spec</link>
<guid isPermaLink="true">https://openai.com/index/updated-model-spec</guid>
<pubDate>Fri, 30 Oct 2026 00:00:00 GMT</pubDate>
<category>OpenAI</category>
<description>OpenAI 更新了 Model Spec，提出五项核心价值观，用于引导能力更强的 AI 走向合理判断、尊重人类自主性并忠实于人类意图。</description>
</item>
<item>
<title>Making it easier for websites to express AI training preferences</title>
<link>https://openai.com/index/helping-websites-express-ai-training-preferences</link>
<guid isPermaLink="true">https://openai.com/index/helping-websites-express-ai-training-preferences</guid>
<pubDate>Fri, 30 Oct 2026 00:00:00 GMT</pubDate>
<category>OpenAI</category>
<description>OpenAI 将遵循共享的 robots.txt 信号，帮助网站运营者向参与该机制的提供商传达其 AI 训练偏好。</description>
</item>
<item>
<title>Fair use benefits everyone</title>
<link>https://openai.com/index/fair-use-benefits-everyone</link>
<guid isPermaLink="true">https://openai.com/index/fair-use-benefits-everyone</guid>
<pubDate>Fri, 30 Oct 2026 00:00:00 GMT</pubDate>
<category>OpenAI</category>
<description>OpenAI 公开了创作者、学者、无障碍倡导者和法律专家等提交的法庭简报节选，内容涉及合理使用如何支持 AI 训练与创新。</description>
</item>
<item>
<title>Community safety report and new protections for teens</title>
<link>https://openai.com/index/community-safety-report-and-new-protections-for-teens</link>
<guid isPermaLink="true">https://openai.com/index/community-safety-report-and-new-protections-for-teens</guid>
<pubDate>Fri, 30 Oct 2026 00:00:00 GMT</pubDate>
<category>OpenAI</category>
<description>OpenAI 发布社区安全报告，并加强 ChatGPT 的暴力预防措施，包括扩展防护机制、更早介入，以及承诺进行公共安全报告。</description>
</item>
<item>
<title>Speeding up planning decisions with AI</title>
<link>https://deepmind.google/blog/speeding-up-planning-decisions-with-ai/</link>
<guid isPermaLink="true">https://deepmind.google/blog/speeding-up-planning-decisions-with-ai/</guid>
<pubDate>Fri, 30 Oct 2026 00:00:00 GMT</pubDate>
<category>Google DeepMind</category>
<description>Google DeepMind 与英国政府合作，缩短规划评估所需时间。</description>
</item>
<item>
<title>Le Chonk: Built for Coders, Trusted by Defenders.</title>
<link>https://mistral.ai/news/mistral-large-4-oss/</link>
<guid isPermaLink="true">https://mistral.ai/news/mistral-large-4-oss/</guid>
<pubDate>Thu, 29 Oct 2026 00:00:00 GMT</pubDate>
<category>Mistral</category>
<description>Mistral Large 4（Le Chonk）的权重已开放。官方称这是一个 1T 参数的多模态模型，也是最佳的开源权重编程模型，在欧洲训练。</description>
</item>
<item>
<title>How we detect frontier model misalignment</title>
<link>https://openai.com/index/how-we-detect-frontier-model-misalignment</link>
<guid isPermaLink="true">https://openai.com/index/how-we-detect-frontier-model-misalignment</guid>
<pubDate>Thu, 29 Oct 2026 00:00:00 GMT</pubDate>
<category>OpenAI</category>
<description>OpenAI 介绍其通过分层监控来检测前沿模型失准和未经授权的智能体行为，以便响应人员在事件升级前介入。</description>
</item>
<item>
<title>Gemini 4 Carbon: Our next frontier model</title>
<link>https://deepmind.google/blog/gemini-4-carbon-our-next-frontier-model/</link>
<guid isPermaLink="true">https://deepmind.google/blog/gemini-4-carbon-our-next-frontier-model/</guid>
<pubDate>Thu, 29 Oct 2026 00:00:00 GMT</pubDate>
<category>Google DeepMind</category>
<description>Google DeepMind 发布 Gemini 4 Carbon，官方称其为下一代前沿模型。</description>
</item>
<item>
<title>Genie 4: A new generation of world simulation</title>
<link>https://deepmind.google/blog/genie-4-a-new-generation-of-world-simulation/</link>
<guid isPermaLink="true">https://deepmind.google/blog/genie-4-a-new-generation-of-world-simulation/</guid>
<pubDate>Wed, 28 Oct 2026 00:00:00 GMT</pubDate>
<category>Google DeepMind</category>
<description>Google DeepMind 发布新一代世界模型 Genie 4，官方称其在实时模拟中实现了前所未有的真实感。</description>
</item>
<item>
<title>OpenAI and ServiceNow expand partnership across enterprise workflows</title>
<link>https://openai.com/index/openai-servicenow-partnership-expansion</link>
<guid isPermaLink="true">https://openai.com/index/openai-servicenow-partnership-expansion</guid>
<pubDate>Wed, 28 Oct 2026 00:00:00 GMT</pubDate>
<category>OpenAI</category>
<description>OpenAI 与 ServiceNow 扩大合作，将 OpenAI 模型和 agent 引入 ServiceNow，并将 ServiceNow 工作流接入 ChatGPT 和 Codex。</description>
</item>
<item>
<title>AI for Science expands to foundations and institutions</title>
<link>https://www.anthropic.com/news/ai-for-science-expands-foundations-research-institutions</link>
<guid isPermaLink="true">https://www.anthropic.com/news/ai-for-science-expands-foundations-research-institutions</guid>
<pubDate>Wed, 28 Oct 2026 00:00:00 GMT</pubDate>
<category>Anthropic</category>
<description>Anthropic 宣布扩展其 AI for Science 项目，并介绍了来自全球的 20 个科学基金会、机构和研究组作为合作伙伴。</description>
</item>
<item>
<title>OpenAI expands partnership with Oracle to bring its apps to ChatGPT</title>
<link>https://openai.com/index/oracle-apps-in-chatgpt</link>
<guid isPermaLink="true">https://openai.com/index/oracle-apps-in-chatgpt</guid>
<pubDate>Tue, 27 Oct 2026 00:00:00 GMT</pubDate>
<category>OpenAI</category>
<description>OpenAI 与 Oracle 扩大合作，将把企业应用和 AI agents 引入 ChatGPT，并为 Oracle 的新产品提供支持。</description>
</item>
<item>
<title>Expanding our partnership with GSA to bring Claude to all levels of government</title>
<link>https://www.anthropic.com/news/expanding-partnership-gsa</link>
<guid isPermaLink="true">https://www.anthropic.com/news/expanding-partnership-gsa</guid>
<pubDate>Tue, 27 Oct 2026 00:00:00 GMT</pubDate>
<category>Anthropic</category>
<description>Anthropic 宣布扩大与美国总务管理局（GSA）的合作，将 Claude 引入各级政府。</description>
</item>
<item>
<title>Anthropic files S-1 with the SEC for proposed initial public offering</title>
<link>https://www.anthropic.com/news/s1-sec</link>
<guid isPermaLink="true">https://www.anthropic.com/news/s1-sec</guid>
<pubDate>Mon, 26 Oct 2026 00:00:00 GMT</pubDate>
<category>Anthropic</category>
<description>Anthropic 向美国证券交易委员会提交 S-1 文件，拟进行首次公开募股。</description>
</item>
<item>
<title>Working with academic communities to build scientific understanding</title>
<link>https://openai.com/index/building-scientific-understanding-with-academic-communities</link>
<guid isPermaLink="true">https://openai.com/index/building-scientific-understanding-with-academic-communities</guid>
<pubDate>Mon, 26 Oct 2026 00:00:00 GMT</pubDate>
<category>OpenAI</category>
<description>OpenAI 公布了模型在 Kakeya 猜想、指数时间假设和量子色动力学上的结果，并说明研究人员如何验证和在此基础上继续研究。</description>
</item>
<item>
<title>ChatGPT’s expanding world of work</title>
<link>https://openai.com/index/chatgpt-expanding-world-of-work</link>
<guid isPermaLink="true">https://openai.com/index/chatgpt-expanding-world-of-work</guid>
<pubDate>Mon, 26 Oct 2026 00:00:00 GMT</pubDate>
<category>OpenAI</category>
<description>OpenAI 发布 Signals 数据，覆盖近 100 个国家、165 项工作活动，展示人们如何用 ChatGPT 处理工作，以及使用方式随时间的变化。</description>
</item>
<item>
<title>Introducing GPT-6.1 Terra</title>
<link>https://openai.com/index/introducing-gpt-6-1-terra</link>
<guid isPermaLink="true">https://openai.com/index/introducing-gpt-6-1-terra</guid>
<pubDate>Mon, 26 Oct 2026 00:00:00 GMT</pubDate>
<category>OpenAI</category>
<description>OpenAI 发布 GPT-6.1 Terra，官方称其为所有人带来快速且强大的计算机使用能力。</description>
</item>
<item>
<title>Rapidly scaling online storage to serve over 1 billion ChatGPT users (Part II)</title>
<link>https://openai.com/index/scaling-storage-one-billion-users-part-two</link>
<guid isPermaLink="true">https://openai.com/index/scaling-storage-one-billion-users-part-two</guid>
<pubDate>Sat, 24 Oct 2026 00:00:00 GMT</pubDate>
<category>OpenAI</category>
<description>OpenAI 介绍如何通过可靠的多租户架构、全局缓存和 Azure Cosmos DB 扩展 Habitat，为 ChatGPT 提供 12 倍更快的读取速度。</description>
</item>
<item>
<title>Build more responsive apps with the Decisions API</title>
<link>https://openai.com/index/decisions-api</link>
<guid isPermaLink="true">https://openai.com/index/decisions-api</guid>
<pubDate>Fri, 23 Oct 2026 00:00:00 GMT</pubDate>
<category>OpenAI</category>
<description>OpenAI 发布 Decisions API，官方称可为任何应用提供快速、聚焦、低成本的决定。</description>
</item>
<item>
<title>OpenAI and Poland’s NASK partner to advance AI safety and security</title>
<link>https://openai.com/index/openai-nask-ai-safety-security-partnership</link>
<guid isPermaLink="true">https://openai.com/index/openai-nask-ai-safety-security-partnership</guid>
<pubDate>Fri, 23 Oct 2026 00:00:00 GMT</pubDate>
<category>OpenAI</category>
<description>OpenAI 与波兰 NASK 合作，围绕 AI 安全、安保与技术研究展开协作，以加强防护措施并支持更具韧性的数字生态。</description>
</item>
<item>
<title>Securing our frontier model training environment</title>
<link>https://openai.com/index/securing-our-frontier-model-training-environment</link>
<guid isPermaLink="true">https://openai.com/index/securing-our-frontier-model-training-environment</guid>
<pubDate>Fri, 23 Oct 2026 00:00:00 GMT</pubDate>
<category>OpenAI</category>
<description>OpenAI 说明其如何加强前沿模型训练与评估的隔离措施，包括离线环境和高算力红队测试。</description>
</item>
<item>
<title>Broad access through consistent standards</title>
<link>https://openai.com/index/broad-access-through-consistent-standards</link>
<guid isPermaLink="true">https://openai.com/index/broad-access-through-consistent-standards</guid>
<pubDate>Thu, 22 Oct 2026 00:00:00 GMT</pubDate>
<category>OpenAI</category>
<description>OpenAI 说明其如何与分发合作伙伴协作，以维持一致的安全保障、支持负责任的使用，并在 Usage Policies 中明确相关期望。</description>
</item>
<item>
<title>Beyond tree cover: mapping the world’s forest types</title>
<link>https://deepmind.google/blog/beyond-tree-cover-mapping-the-worlds-forest-types/</link>
<guid isPermaLink="true">https://deepmind.google/blog/beyond-tree-cover-mapping-the-worlds-forest-types/</guid>
<pubDate>Thu, 22 Oct 2026 00:00:00 GMT</pubDate>
<category>Google DeepMind</category>
<description>Google DeepMind 发布了一张全球地图，将地球上每一块 10 米见方的土地归类为不同的森林类型。</description>
</item>
<item>
<title>Open call for guest investigators in the life sciences</title>
<link>https://www.anthropic.com/news/open-call-guest-investigators-life-sciences</link>
<guid isPermaLink="true">https://www.anthropic.com/news/open-call-guest-investigators-life-sciences</guid>
<pubDate>Thu, 22 Oct 2026 00:00:00 GMT</pubDate>
<category>Anthropic</category>
<description>Anthropic 邀请生命科学界人士申请参与特别项目。</description>
</item>
<item>
<title>Southeast Asia’s digital economy to hit $368 billion</title>
<link>https://blog.google/company-news/inside-google/around-the-globe/google-asia/e-conomy-sea-report-2026/</link>
<guid isPermaLink="true">https://blog.google/company-news/inside-google/around-the-globe/google-asia/e-conomy-sea-report-2026/</guid>
<pubDate>Wed, 21 Oct 2026 00:00:00 GMT</pubDate>
<category>Google AI</category>
<description>AI 的广泛采用正推动东南亚数字经济在 2026 年达到 3680 亿美元，并在 2035 年达到 1 万亿美元。</description>
</item>
<item>
<title>OpenAI models are coming to Google Cloud</title>
<link>https://openai.com/index/openai-models-google-cloud</link>
<guid isPermaLink="true">https://openai.com/index/openai-models-google-cloud</guid>
<pubDate>Wed, 21 Oct 2026 00:00:00 GMT</pubDate>
<category>OpenAI</category>
<description>OpenAI 与 Google Cloud 将把前沿模型和 Codex 工作流引入 Gemini 的 Model Garden，并通过 Google Cloud Marketplace 提供。</description>
</item>
<item>
<title>Using thousands of AI agents to uncover new leads in biological data</title>
<link>https://openai.com/index/scaling-ai-research-for-complex-biological-data</link>
<guid isPermaLink="true">https://openai.com/index/scaling-ai-research-for-complex-biological-data</guid>
<pubDate>Wed, 21 Oct 2026 00:00:00 GMT</pubDate>
<category>OpenAI</category>
<description>OpenAI、PerturbAI 和 Scripps Research 使用分布式 AI agents，从数百万个脑细胞的数据中发现了有前景的研究线索。</description>
</item>
<item>
<title>How we built ChatGPT’s novel ad stack for user trust and relevance</title>
<link>https://openai.com/index/building-chatgpt-ad-stack</link>
<guid isPermaLink="true">https://openai.com/index/building-chatgpt-ad-stack</guid>
<pubDate>Wed, 21 Oct 2026 00:00:00 GMT</pubDate>
<category>OpenAI</category>
<description>OpenAI 介绍 ChatGPT 广告系统的构建方式：该系统由 LLM 驱动，用于理解对话意图，将广告与回答分开，并大规模投放相关广告。</description>
</item>
<item>
<title>Mistral Advances AI for Engineering with NVIDIA</title>
<link>https://mistral.ai/news/mistral-nvidia-ai-for-industrial-engineering/</link>
<guid isPermaLink="true">https://mistral.ai/news/mistral-nvidia-ai-for-industrial-engineering/</guid>
<pubDate>Wed, 21 Oct 2026 00:00:00 GMT</pubDate>
<category>Mistral</category>
<description>Mistral 与 NVIDIA 合作，基于 NVIDIA 加速计算开发新一代 AI 原生工程工作流。</description>
</item>
<item>
<title>Patch the Planet: global response coordination as agents uncover critical vulnerabilities</title>
<link>https://openai.com/index/coordinated-disclosure-of-critical-vulnerabilities</link>
<guid isPermaLink="true">https://openai.com/index/coordinated-disclosure-of-critical-vulnerabilities</guid>
<pubDate>Wed, 21 Oct 2026 00:00:00 GMT</pubDate>
<category>OpenAI</category>
<description>OpenAI 介绍 Patch the Planet，该项目连接安全研究人员、维护者和事件响应者，帮助处理开源软件中的关键漏洞。</description>
</item>
<item>
<title>Democratic governance of frontier AI, part II: Empowering cyber defenders</title>
<link>https://openai.com/index/empowering-cyber-defenders</link>
<guid isPermaLink="true">https://openai.com/index/empowering-cyber-defenders</guid>
<pubDate>Wed, 21 Oct 2026 00:00:00 GMT</pubDate>
<category>OpenAI</category>
<description>OpenAI 提出一份蓝图，内容涉及用前沿 AI 赋能网络防御者、加强防护措施、推进网络安全现代化，以及提升公共机构的韧性。</description>
</item>
<item>
<title>Physicist Christopher Stubbs joins Anthropic&amp;#x27;s Long-Term Benefit Trust</title>
<link>https://www.anthropic.com/news/christopher-stubbs-joins-long-term-benefit-trust</link>
<guid isPermaLink="true">https://www.anthropic.com/news/christopher-stubbs-joins-long-term-benefit-trust</guid>
<pubDate>Wed, 21 Oct 2026 00:00:00 GMT</pubDate>
<category>Anthropic</category>
<description>哈佛大学实验物理学家、前理学院院长 Christopher Stubbs 加入 Anthropic 的 Long-Term Benefit Trust。他长期从事观测宇宙学与精密测量研究，并曾就国家安全与军控问题为美国政府提供咨询。Stubbs 表示，他每天在研究中使用 AI 工具，作为教育者和祖父，希望帮助确保 AI 的开发和部署造福人类。</description>
</item>
</channel>
</rss>
