[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"news-slug-runway-video-world-model-google-bet":3,"news-related-b7e375b3-6949-4ff6-9cd4-012aced6bbc3":36},{"id":4,"title":5,"summary":6,"content":6,"original_url":7,"source_id":8,"tags":9,"translations":23,"news_slug":29,"published_at":30,"created_at":31,"modified_at":32,"is_published":33,"publish_type":34,"image_url":13,"view_count":35},"b7e375b3-6949-4ff6-9cd4-012aced6bbc3","Runway赌注视频生成通往世界模型：挑战Google的架构之赌","当整个AI行业押注语言模型时，总部位于纽约的创业公司Runway选择了另一条路——从视频数据而非文本中构建智能。\n\nRunway联合创始人Anastasis Germanidis的观点颇具挑衅性：「语言模型在互联网、论坛、社交媒体和教科书上训练——是在蒸馏已有的人类知识。但要超越它，我们需要利用偏见更少的数据。」这句话背后是世界模型的野心：通过视频直接观察世界运作，而非通过人类语言间接描述。\n\nRunway成立于2018年，三位创始人均来自纽约大学Tisch艺术学院，而非典型的硅谷背景。他们从视频生成起家，Gen-4.5已是第四代模型，服务于狮门影业、AMC Networks等头部客户，并被用于《瞬息全宇宙》等电影制作。公司估值已达53亿美元，2026年第二季度新增4000万美元年经常性收入。\n\n然而真正的赌注在于架构选择。当GPT-5.5和Claude Opus 4.7在语言基准上你追我赶时，Runway认为下一代智能将来自视频而非文本。如果这一判断正确，Google不会是唯一需要担心的对手——整个以语言为基础的AI路线都可能面临重估。\n\n这并非没有风险。Google拥有更深的口袋和更多的数据储备。但Germanidis相信，正是因为Runway不是来自Google或Meta，才有可能做出不同的赌注。AI竞争的下半场，或许不仅仅是参数量和基准分数的竞争，而是关于「智能建立在什么数据之上」的根本分歧。","https:\u002F\u002Ftechcrunch.com\u002F2026\u002F05\u002F15\u002Frunway-started-by-helping-filmmakers-now-it-wants-to-beat-google-at-ai\u002F","226bcb3d-18b8-4bb0-a999-4e82ec13f5fd",[10,14,17,20],{"id":11,"name":12,"slug":12,"description":13,"color":13},"e676a5cf-1f24-472f-a765-86fa21a1bc3c","ai-model",null,{"id":15,"name":16,"slug":16,"description":13,"color":13},"40269b40-7942-4650-9672-ed2e6524d37a","ai-technology",{"id":18,"name":19,"slug":19,"description":13,"color":13},"499f4b56-819d-49a3-9609-33e775143b86","multimodal",{"id":21,"name":22,"slug":22,"description":13,"color":13},"ebe5dcd1-46b1-4298-b8c2-8e0e2f456e56","video-generation",[24],{"id":25,"lang":26,"title":27,"summary":28,"content":13},"a5a966ee-e1ed-4ac0-b6e5-dfced99d1456","en","Runway bets video generation becomes a world model","As the entire AI industry bets on language models, New York-based startup Runway has chosen another path — building intelligence from video data rather than text.\n\nRunway co-founder Anastasis Germanidis offers a provocative take: \"Language models are trained on the internet, forums, social media, and textbooks — distilling existing human knowledge. But to surpass them, we need to leverage data with fewer biases.\" This statement underpins a world-model ambition: observing the world operate directly through video, rather than indirectly describing it through human language.\n\nRunway was founded in 2018, with all three co-founders from NYU's Tisch School of the Arts, not the typical Silicon Valley background. They started in video generation, and Gen-4.5 is their fourth-generation model, serving top-tier clients like Lionsgate and AMC Networks, and has been used in productions like \"Everything Everywhere All at Once.\" The company is valued at $5.3 billion, with $40 million in new annual recurring revenue added in Q2 2026.\n\nBut the real bet is the architectural choice. While GPT-5.5 and Claude Opus 4.7 chase each other on language benchmarks, Runway believes the next generation of intelligence will come from video, not text. If this judgment is correct, Google won't be the only opponent that needs to worry — the entire language-based AI route could face reevaluation.\n\nThis isn't without risks. Google has deeper pockets and more data reserves. But Germanidis believes that precisely because Runway isn't from Google or Meta, it's possible to make a different bet. The second half of the AI competition may not be just about parameter counts and benchmark scores, but a fundamental divide over \"what data intelligence is built upon.\"","runway-video-world-model-google-bet","2026-05-16T04:10:00Z","2026-05-16T04:06:44.922589Z","2026-08-19T02:08:40.142862Z",true,"agent",88,{"items":37},[38,43,48,53,58,63],{"id":39,"title":40,"news_slug":41,"published_at":42},"ff7fa3e1-6737-4bd9-85c8-8010d13a44f3","字节跳动 Bernini 开源：用 MLLM 当\"语义规划师\"，拆开视频生成的\"思考\"与\"渲染\"","bytedance-bernini-mllm-semantic-planner-diT","2026-05-21T00:00:00+00:00",{"id":44,"title":45,"news_slug":46,"published_at":47},"faab4a6c-9cb0-4f2a-a5bf-1f122306008b","Wan-Streamer v0.2：分辨率 192p→640p，保住 200ms","alibaba-wan-streamer-v0-2","2026-07-10T16:15:00+00:00",{"id":49,"title":50,"news_slug":51,"published_at":52},"aeb60d9a-6639-4669-96a4-951aadad40cb","AI 视频工具进入「全场景」分化期:6 款主流产品的技术路线对比","ai-video-tools-comparison","2026-07-08T08:00:00+00:00",{"id":54,"title":55,"news_slug":56,"published_at":57},"a2e8ac5b-ca51-4ddb-88d4-54373d1f0774","SUNTA 用\"惊奇度\"切分视频预测:东京大学让模型在 250 步后仍不崩溃","sunta-surprise-chunking-video","2026-07-04T16:00:00+00:00",{"id":59,"title":60,"news_slug":61,"published_at":62},"2657cbe0-7743-43f2-9332-ee18b84b1229","Directing the World: 中国电信 TeleAI 把自回归视频世界模型推到\"组合控制\"","teleai-directing-the-world","2026-07-01T10:30:00+00:00",{"id":64,"title":65,"news_slug":66,"published_at":67},"efa5f558-e27f-4889-bf71-74cbef874ace","即梦 AI 把 Seedance 2.0 推上原生 4K:把超分这道工序从后期流水线里拿掉","seedance-2-0-native-4k-jimeng","2026-06-24T08:00:00+00:00"]