[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"$f_f19U2CzxF6KO1FmZ_hdiF65ChaPYmWAfI3rdCMzOMs":3},{"item":4,"related":52},{"id":5,"type":6,"title":7,"slug":8,"summary":9,"body":10,"coverUrl":11,"productScreenshots":12,"productLinks":13,"authorName":14,"authorUrl":15,"authorSubject":16,"category":17,"tags":22,"sourceLabel":43,"sourceName":44,"sourceUrl":44,"status":45,"seoTitle":44,"seoDescription":44,"canonicalUrl":44,"isFeatured":46,"sno":47,"sortOrder":48,"publishedAt":49,"updatedAt":50,"createdAt":51},"b0a1b579-5f92-4b34-b6a5-27f50fa6b3d4","article","Jev：一款不擅长聊天的AI，为什么可能改变未来的游戏和互联网？","jev","它不擅长聊天，也不会直接生成长篇文章，而是专注于快速、低成本地作出判断。","近年来，人工智能的发展大多围绕一个目标：让AI更聪明、更全能。从ChatGPT到各种多模态模型，AI已经能写文章、编代码、生成图片，甚至完成复杂推理。\n \nJev却走了另一条路，它不擅长聊天，也不写长文，而是专注快速、低成本地作出判断。\n\n## Jev有什么不同？\n\n2026年9月15日，TypeSafe AI推出Jev。创始人Diogo Almeida曾参与OpenAI研究。TypeSafe AI称其为首款“System One Model”（系统一模型），灵感来自心理学中的快速、直觉式思维。传统大语言模型生成文字，Jev负责结构化决策。。\n\n## 让判断变得更便宜\n\nJev的响应时间约70至500毫秒，每百万输入Token收费0.042美元，输出不另收费。单次调用节省有限，但频繁调用时差别明显。\n\nAI智能体常要判断：是否继续搜索、调用哪个工具、信息是否足够、何时转人工。把这些简单决策交给Jev，复杂推理和内容生成留给大模型，可降低延迟和成本。\n\n> 过去因成本过高而难以落地的AI应用，因此有了落地空间。\n\n## 从实时游戏到动态网页\n\n### 实时生成游戏关卡\n\n2026年9月18日，开发者Hugo Duprez展示用Jev实时生成跑酷关卡。系统把玩家位置、速度和地形发给Jev，由它决定前方道路的宽度、间隔、高度和地形类型。演示中，单次响应约319至375毫秒，五次请求成本约0.00286美元。\n\n\u003Cimg src=\"\u002Fuploads\u002F2026-09-29\u002F543031d5-372f-4dcc-a54f-13136ec2ba43.jpg\" alt=\"IMG_7679\">\n\n游戏可根据玩家水平、战斗方式或选择，动态调整关卡和任务。Jev不生成完整画面，只决定“生成什么”，具体内容仍由游戏引擎和素材库完成。\n\n### 动态网页\n\nVercel Labs在开源项目json-render中实验：让Jev从预设组件中选择、排列、分组，再组合成界面。功能仍处于实验阶段。\n\nhttps:\u002F\u002Fjson-render.dev\u002F\n\n比如买电脑时优先展示性能对比和价格；申请售后时突出订单、退换货和客服入口。教育、办公软件也可按需调整页面。\n\n\u003Cimg src=\"\u002Fuploads\u002F2026-09-29\u002F54685744-ea13-4c11-a9d1-b6ed06d10f92.jpg\" alt=\"b3358ac6-359b-49cb-bd70-fd2dfefdac7d\">\n\n网页不再只有一套固定界面，而是能根据用户需求变化。\n\n但Jev目前只能从已有组件中选择组合，不能凭空创造内容。\n\n## 让软件走向动态智能\n\n过去软件靠提前写好的规则运行。Jev这类模型，让软件能根据具体情境灵活判断。开发者定义功能、内容和安全边界，AI根据用户行为和环境变化决定如何调用。\n\n智能家居可调整运行方式，教育软件可改变教学节奏，AI智能体可随任务进展选择工具。\n\n> AI不再只在用户提问时工作，而是融入软件日常运行，在后台持续决策。\n\nTypeSafe AI将Jev命名为Jevons缩写，受“杰文斯悖论”启发：资源使用效率提高、成本降低，人们反而用得更多。智能判断足够便宜时，AI就可能进入过去不值得用的环节。Jev的市场空间，可能不只是AI分类和自动化，还包括实时交互软件生态。\n\n## AI的下一步，可能不只是变得更聪明\n\nJev提供了另一种思路：AI不必都追求更大、更强，也可以通过专业分工，让智能融入软件运行。它未必取代ChatGPT，但可能与其互补。ChatGPT让人们看到AI能创造什么，Jev这样的模型，则在探索AI如何持续决定接下来发生什么。\n\n未来软件也许不再只按固定规则运行，而是能在安全可控范围内，根据需求不断调整功能和行为。","\u002Fuploads\u002F2026-09-28\u002F5636be82-4042-4a65-be02-a58a5496cdca.jpg",[],[],"Foundit","https:\u002F\u002Ffoundit.cn","f39339b1-aaa6-4e86-b0c2-a6e6a21113b5",{"id":18,"name":19,"slug":20,"description":21},"6179d3b6-dc34-4483-9ded-3cd9f1b37a47","科普","abbreviation","介绍各领域新兴概念",[23,27,31,35,39],{"id":24,"name":25,"slug":26},"88d2bc27-0e0f-468a-b907-2991cb97b87b","人工智能","ai",{"id":28,"name":29,"slug":30},"7c76bfc2-f80f-4ee0-a95d-27bd8708b434","技术","slug",{"id":32,"name":33,"slug":34},"4c2bbea6-eab7-40a8-8447-1de478ff7749","分析","analyse",{"id":36,"name":37,"slug":38},"82f427f8-6275-4cbb-bcac-cf1948488006","网页应用","web-app",{"id":40,"name":41,"slug":42},"1404d044-7b7e-4cdc-b703-a5e6df8fda50","游戏","game","资料来源",null,"published",true,1,0,"2026-09-28T00:00:00.000Z","2026-09-29T02:58:12.133Z","2026-09-27T13:08:43.665Z",[53,61,70],{"id":54,"type":6,"title":55,"slug":56,"summary":57,"coverUrl":58,"authorName":14,"sno":47,"publishedAt":59,"createdAt":60},"bccd34d2-d0a9-4f24-9f3e-ba93b6e8169f","Intelligent UI：AI正在重新定义软件交互方式","intelligent-ui","从固定界面走向动态生成，从功能驱动走向意图驱动，Intelligent UI所代表的，可能正是下一代智能软件的重要发展方向。","\u002Fuploads\u002F2026-10-08\u002Fb73ef614-9f98-4cee-936f-e5b507cd9f94.jpg","2026-10-08T00:00:00.000Z","2026-10-08T06:16:19.878Z",{"id":62,"type":6,"title":63,"slug":64,"summary":65,"coverUrl":66,"authorName":14,"sno":67,"publishedAt":68,"createdAt":69},"872963cd-242c-4912-a3dc-9497386d4772","Certificate Transparency：TLS 证书为什么要公开留日志？","certificate-transparency-public-audit-logs","证书透明度通过公开、可验证的日志提升 TLS 证书签发的可观察性。本文说明监控者如何发现意外证书，也厘清 CT 不会自动阻止误发。","\u002Fuploads\u002F2026-10-06\u002Fa2b8f9d0-fb90-4f19-ae2b-22cdebd395f2.jpg",43,"2026-10-06T00:00:00.000Z","2026-10-06T07:51:39.208Z",{"id":71,"type":6,"title":72,"slug":73,"summary":74,"coverUrl":75,"authorName":14,"sno":76,"publishedAt":77,"createdAt":78},"3eee805c-a6c6-438b-b41c-bb67359211f6","HNSW：向量数据库怎样在“小世界图”里快速找相似项？","hnsw-approximate-nearest-neighbor-search","HNSW 把向量组织为多层可导航图，从稀疏高层逐步下探到局部候选区域，以较少距离计算近似查找邻居。本文讲解图搜索原理，以及召回率、延迟、构建和内存之间的权衡。","\u002Fuploads\u002F2026-10-13\u002Feb9f9a0c-fcbb-4323-86ea-58d5ca17c9f2.jpg",47,"2026-10-13T00:00:00.000Z","2026-10-13T02:14:06.187Z"]