《纽约时报》周一起诉 Brave Software，指控这家浏览器制造商抓取其新闻内容，并将访问权限出售给 AI 公司。
The New York Times sued Brave Software on Monday, accusing the browser maker of scraping its journalism and selling access to it to AI companies.
数小时后，《纽约每日新闻》、《芝加哥论坛报》和《奥兰多哨兵报》的出版商提起了另一项诉讼，提出类似指控。
Hours later, the publishers of the New York Daily News, the Chicago Tribune and the Orlando Sentinel filed a separate lawsuit making similar allegations.
这两起案件均在曼哈顿联邦法院提起，且都聚焦于 Brave 业务的同一部分：其搜索索引以及向 AI 开发者提供的数据。
Both cases were brought in federal court in Manhattan, and both focus on the same part of Brave’s business: its search index and the data it makes available to AI developers.
三家报纸在其起诉状中称，Brave 的行为“就像一个窃贼，挪用出版商的内容，又像一个 AI ‘赃车贩子’，出售被盗内容”，我们已审阅了该起诉状。
Brave acts like “a burglar that misappropriates the content of publishers, and an AI ‘fence’ that sells stolen content,” the three newspapers say in their complaint, which we have reviewed.
MediaNews Group 和 Tribune Publishing 的律师 Steven Lieberman 在声明中说：“AI 巨头现在拼命想避免被当场抓获、从出版商那里窃取新闻，因此 Brave Software 正向它们出售一条后门，通往它们明知无权获取的新闻内容。购买二手赃物并不能使其合法化。Brave 必须为向 AI 行业‘洗白’盗版报道承担责任。”《纽约时报》发言人 Graham James 告诉路透社，该公司曾多次要求 Brave 停止使用其新闻内容，但这家浏览器制造商置之不理。
Steven Lieberman, attorney for the MediaNews Group and Tribune Publishing, said in a statement: “AI giants are now desperate to avoid getting caught red-handed stealing news from publishers, so Brave Software is selling them a back door to the journalism they know they have no right to. Buying stolen goods secondhand doesn’t make them legitimate. Brave must be held responsible for laundering pirated reporting for the AI industry.”
三家报纸提出了五项诉由：直接、帮助和替代性版权侵权，以及根据《数字千年版权法》规避技术保护措施。《纽约时报》则根据 MLex 的报道，提起了自己的版权和 DMCA 诉由。
New York Times spokesperson Graham James told Reuters that the company had repeatedly asked Brave to stop using its journalism, but the browser maker had ignored those requests. The three newspapers have brought five claims: direct, contributory and vicarious copyright infringement, circumvention of technical protections under the Digital Millennium The Times has filed its own copyright and DMCA claims, according to MLex.
两起诉讼的核心都是 Brave 索引，这是一个包含 300 多亿网页的集合，为 Brave Search 提供支持。出版商称，Brave 从未披露其用于收集材料的爬虫名称。
At the heart of both lawsuits is the Brave Index, a collection of more than 30 billion web pages that powers Brave Search. The publishers say Brave has never disclosed the name of the crawler it uses to collect the material.
这很重要，因为网站可以按名称屏蔽自动爬虫，如果机器人的身份未知，出版商就无法识别并排除它。
That matters because websites can block automated crawlers by name, leaving publishers unable to identify and exclude a bot whose identity is unknown.
Brave 随后通过其搜索 API 出售对该索引的访问权限。两份起诉书均指认了相同的五家客户：Anthropic、Perplexity、Mistral、xAI 和 Cohere。Brave 自身的营销材料将该 API 最常见的用途描述为“用于训练和奠定网络上几个最大的大型语言模型的基础”。报社还引用了一篇 Brave 博客文章，称 Anthropic 曾使用该 API 训练 Claude 2，当时这家 AI 公司正准备进行一场大型首次公开募股（IPO）。
Brave then sells access to the index through its Search API. Both complaints identify the same five customers: Anthropic, Perplexity, Mistral, xAI and Cohere. Brave’s own marketing describes the API’s most common uses as “in training and grounding several of the Web’s biggest large language models.”The newspapers also cite a Brave blog post stating that Anthropic used the API to train Claude 2, in the context where the AI company is now preparing for a major IPO.
起诉指控不仅限于提供给其他 AI 公司的数据。出版商辩称，Brave 自身的 AI 搜索功能同样复制了他们的作品。据称，Brave Search 中的 AI 答题功能 Ask Brave 会逐字逐句返回段落，并生成其文章的详细摘要。
The complaints extend beyond the data supplied to other AI companies. The publishers argue that Brave’s own AI search features reproduce their work, too. Ask Brave, the AI answer feature in Brave Search, allegedly returns passages word for word and produces detailed summaries of their articles.
另一个攻击目标是一项名为“付费墙拦截器”的浏览器设置，Brave 将其宣传为绕过付费墙的一种方式。这三家报纸称，该功能允许用户在无需登录的情况下访问订阅内容。
Another target is a browser setting called Paywall blocker, which Brave promotes as a way to bypass paywalls. The three newspapers say it allows users to access subscription content without logging in.
起诉书引用了一位 Reddit 用户的说法，该用户开启该功能后发现，可以“不受干扰且无需设置账户”地阅读《纽约时报》。报社还质疑 Brave 如何通过广告获利。其浏览器默认拦截出版商的广告，同时允许 Brave 销售自己的广告。这三家由 Alden 拥有的出版物辩称，这让该公司能够从新闻业中获利，而无需与生产新闻的出版商分享收入。
Their complaint quotes a Reddit user who switched on the feature and found they could read the New York Times “without interruption and without needing to set up an account.”The newspapers also challenge how Brave makes money from advertising. Its browser blocks publishers’ ads by default while allowing Brave to sell advertising of its own. The three Alden-owned publications argue that this lets the company profit from journalism without sharing the revenue with the publishers who produce it.
法律诉求背后存在一个更广泛的担忧：出版商担心 AI 驱动的搜索正在将读者从他们的网站上拉走。
There is a broader concern behind the legal claims: publishers fear that AI-powered search is taking readers away from their websites.
这三家报纸引用了一项研究，发现 AI 爬虫产生的推荐流量比传统的 Google 搜索少 95.7%，导致出版商将新闻转化为广告和订阅收入的机会减少。
The three newspapers cite a study finding that AI crawlers generate 95.7% less referral traffic than traditional Google search, leaving publishers with fewer opportunities to turn their journalism into advertising and subscription revenue.
奥尔登旗下报纸称，超过200万篇已注册的文章受到影响。它们要求获得法定赔偿和实际损害赔偿、返还利润以及永久禁令，并申请了陪审团审判。
The Alden-owned papers say more than two million registered articles are affected. They are seeking statutory and actual damages, repayment of profits and a permanent injunction, and have requested a jury trial.
根据报纸的起诉书，Brave公司2025年收入超过1亿美元，拥有约1.3亿月活用户。
Brave reported more than $100m in revenue in 2025 and has around 130 million monthly users, according to the newspapers’ complaint.
该公司此前已因其索引做法面临法律诉讼。2025年3月，Brave起诉新闻集团，请求法院宣告其网页索引属于合理使用。新闻集团于7月提起反诉，指控Brave盗窃。
The company has already faced a legal fight over its indexing practices. In March 2025, Brave sued News Corp, asking a court to declare its web indexing fair use. News Corp countersued in July, accusing Brave of theft.
《纽约时报》和奥尔登旗下报纸还分别对OpenAI和微软提起了版权诉讼。
The Times and the Alden-owned newspapers have also filed separate copyright lawsuits against OpenAI and Microsoft.