CNN sues Perplexity over ‘verbatim’ copycat articles CNN起诉Perplexity“逐字抄袭”文章

The Legal Foundation: A Copyright Infringement Claim

The lawsuit is fundamentally a copyright infringement case. CNN's central argument is that Perplexity's AI "answer" engine and browser, Comet, reproduce protected works. The filing specifies that the AI tools generate "verbatim" copies, which moves the accusation beyond mere summarization or fair use. By doing so, CNN positions the issue not as a transformative use but as direct, unauthorized copying of its proprietary content.

Perplexity's Alleged Operational Model: Ignoring Technical Barriers

A key allegation is that Perplexity actively circumvents technical measures designed to control access. The lawsuit states the company ignored CNN's efforts "to recognize or block Perplexity's unidentified crawlers." This suggests Perplexity is accused of deliberately employing scrapers that do not identify themselves or respect standard blocking protocols like robots.txt. Such behavior, if proven, would undermine a common method websites use to manage bot traffic and protect content, escalating the dispute beyond simple data collection to intentional evasion.

The Impact on Content Creation and Monetization

CNN frames the harm as a direct threat to its business model and the human labor behind its product. The claim that Perplexity provides users with subscription-locked information without compensation directly challenges CNN's revenue stream. Furthermore, the lawsuit emphasizes the human effort in reporting and writing, arguing that Perplexity's model "takes without permission or compensation." This highlights a core tension: AI models that train on and regurgitate existing content can potentially devalue the original work that creates that content, jeopardizing the economic ecosystem for professional journalism.

Broader Implications for AI and Content Scraping

This case exemplifies a growing conflict between AI developers and content creators. The outcome could set significant precedent for how copyright law applies to AI training data and output generation. If courts find that verbatim reproduction or paywall circumvention by AI constitutes infringement, it could force AI companies to develop more rigorous content sourcing agreements or fundamentally alter how their models interact with published works. The suit challenges the notion that all data on the web is freely available for AI training, emphasizing the importance of respecting both legal rights and technical boundaries.

诉讼的核心指控：超越“训练”范畴的即时侵权

此次诉讼的焦点并非传统意义上使用新闻内容训练AI模型，而是指向Perplexity的即时产品功能。指控明确区分于数据挖掘或模型训练环节，直指其在向用户提供答案时：

• 生成“逐字复制”（verbatim）的内容，构成直接复制。
• 提供信息的渠道绕过了CNN的订阅付费墙，直接损害了其商业订阅模式。
• 无视CNN为阻止其爬虫所做的技术努力，包括针对“未识别爬虫”的封锁。

这表明，指控的核心是Perplexity的AI作为一个实时的、侵权性的内容分发工具，而非仅仅是数据的使用者。

AI内容抓取的灰色地带正在收窄

传统上，互联网允许有限度的爬虫抓取以索引网页。但AI公司，特别是提供“答案引擎”或“AI搜索”的Perplexity，其爬虫行为的本质引发了新的法律争议。

• “合理使用”边界被挑战：直接复制并替代原文访问（尤其是付费内容），很难符合版权法中关于“转换性使用”的庇护。
• 技术对抗升级：CNN指出Perplexity使用“未识别的爬虫”，这本身可能是规避网站许可协议或robots.txt等标准控制文件的技术手段。诉讼将这种技术规避行为本身作为侵权故意的证据之一。

诉讼预示的行业风向：从“用内容训练”到“用内容直接服务”

此案可能标志着内容行业与AI公司斗争的一个新阶段。争议焦点从“AI是否可以使用我们的内容进行学习？”转向了更尖锐的问题：“AI是否可以直接复制并转售我们的内容服务？”

• 商业模式的根本冲突：CNN的核心资产是其高质量的人工生产内容，并通过订阅墙变现。Perplexity的模式若被法律认可，将直接掏空这一商业模式的基础。
• 对AI爬虫规范化的压力：无论诉讼结果如何，此案都会迫使AI公司重新评估其内容抓取策略，未来与出版商之间更正式、更透明的许可协议框架或将加速形成。