内容来源：https://techcrunch.com/2025/12/04/chicago-tribune-sues-perplexity/

内容总结：

美国《芝加哥论坛报》于当地时间周四正式对人工智能搜索引擎Perplexity提起版权侵权诉讼，案件已由纽约联邦法院受理。这是近期新闻出版业与人工智能公司之间一系列法律纠纷的最新进展。

诉状显示，《芝加哥论坛报》律师于十月中旬曾致函Perplexity，询问其是否使用该报内容。Perplexity方面回复称未使用该报作品训练模型，但承认“可能接收非逐字摘录的事实性摘要”。报社律师对此提出反驳，指控Perplexity实际在提供逐字转载的原文内容。

值得关注的是，本案首次将“检索增强生成”（RAG）技术列为指控焦点。该技术本是为确保信息准确性而设计，通过调用已验证数据源来减少人工智能“幻觉”。但诉状指出，Perplexity未经授权擅自抓取报社内容用于RAG系统，其Comet浏览器更绕开付费墙直接生成文章详细摘要。

此次诉讼是新闻出版业维护版权的最新行动。今年4月，包括《芝加哥论坛报》在内的17家媒体集团旗下机构已就训练数据问题起诉OpenAI和微软；11月另有9家出版机构提起类似诉讼。

除本案外，Perplexity还面临多起诉讼：Reddit于10月提起诉讼，道琼斯公司也已采取法律行动。上月亚马逊虽未正式起诉，但已就AI浏览器购物功能向其发出禁止函。

目前Perplexity未对本次诉讼及相关问询作出回应。随着创作者与人工智能公司的法律博弈持续升级，司法系统是否将对RAG技术应用的法律责任作出界定，将成为行业关注的重要风向标。

中文翻译：

《芝加哥论坛报》周四对人工智能搜索引擎Perplexity提起诉讼，指控其侵犯版权。TechCrunch获悉的诉状已提交至纽约联邦法院。

诉状显示，《芝加哥论坛报》方面称其律师于十月中旬联系Perplexity，询问该AI搜索引擎是否使用其内容。据诉讼文件披露，Perplexity的律师回复称并未使用论坛报的内容训练模型，但承认"可能接收非逐字的事实性摘要"。

然而论坛报的律师指出，Perplexity实际在逐字传播其受版权保护的内容。值得注意的是，报社律师还将矛头指向Perplexity采用的检索增强生成技术。该技术本应通过限定模型仅使用准确可靠的资料来源来减少事实错误，但论坛报指控Perplexity未经许可抓取其内容用于该技术系统。诉状还指出，Perplexity的Comet浏览器甚至绕过报社付费墙，直接生成文章的详细摘要。

今年四月，论坛报所属的媒体新闻集团与论坛报出版公司旗下共17家新闻机构，就模型训练资料问题起诉了OpenAI和微软，该案仍在审理中。去年十一月，这两家出版集团旗下另有九家机构也对相关模型开发商及其云服务提供商提起了诉讼。

尽管创作者已就模型训练数据问题对AI公司提起多起诉讼，但法院是否会就检索增强生成技术的法律责任作出裁决仍有待观察。Perplexity未立即回应《芝加哥论坛报》对其涉诉报道的询问，也未回复TechCrunch的置评请求。目前该公司还面临其他类似诉讼：Reddit于去年十月提起诉讼，道琼斯公司也已提出告诉。上月亚马逊虽未正式起诉，但已就AI浏览器购物功能发出侵权警告函。

英文来源：

The Chicago Tribune filed a lawsuit against AI search engine Perplexity on Thursday alleging copyright infringement. The suit, seen by TechCrunch, was filed in a federal court in New York.
The Tribune alleges that its lawyers contacted Perplexity in mid-October asking if the AI search engine was using its content, according to the complaint. Perplexity’s lawyers replied it did not train models with the Tribune’s work, but that it “may receive non-verbatim factual summaries,” the lawsuit claims.
The Tribune’s lawyers, however, argue that Perplexity is delivering Tribune content verbatim.
Interestingly, the newspaper’s lawyers are also calling out Perplexity’s Retrieval Augmented Generation (RAG) as a culprit. RAG is a method used to limit hallucinations by having the model only use an accurate or verified data source. The Tribune argues that Perplexity is using the newspaper’s content in its RAG systems, scraped without permission. Plus, it alleges the Perplexity’s Comet browser is bypassing the paper’s paywall to deliver detailed summaries of those articles.
The Tribune is one of 17 news publications from MediaNews Group and Tribune Publishing that sued OpenAI and Microsoft over model training material in April. That suit is ongoing. Another nine from these publishers sued the model maker and its cloud provider in November, too.
While creators have filed many lawsuits against model makers over using their work for model training, we’ll have to see if the courts weigh in about the legal liabilities of RAG as well.
Perplexity did not immediately respond to the Chicago Tribune’s story about its own lawsuit, nor to TechCrunch’s request for comment. Perplexity is facing other such suits. Reddit filed one in October. Dow Jones is also suing. Last month, while Amazon didn’t sue, it did threaten to by sending a cease and desist letter over AI browser shopping.