We are tracking a significant legal development in the federal court system where CNN has filed a formal complaint against Perplexity, an AI-assisted search engine. The lawsuit centers on allegations that Perplexity’s business model depends on the systematic misappropriation of copyrighted news content and the unauthorized use of trademarks. This case is a major test for how traditional media protections apply to generative AI platforms.
Understanding the Basis of the CNN Lawsuit Against Perplexity
Current intellectual property laws exist to ensure that creators maintain exclusive rights to their original works. Under the Copyright Act, we look at whether a secondary user is creating a “derivative work” or simply copying a substantial portion of the original without payment. If a platform reproduces enough content to serve as a market substitute for the original, it typically violates the owner’s rights. Trademark law operates on a different principle called “likelihood of confusion.” It prevents a company from using another brand’s name or logo if that use makes consumers believe the brand endorsed or produced the second company’s output.
Analyzing the Copyright Infringement Claims in AI Content Scraping
CNN alleges that Perplexity utilizes automated crawlers to ingest vast amounts of journalism. And while Perplexity markets itself as an “answer engine,” the lawsuit argues it is actually a “content recycler.” The legal filing asserts that Perplexity’s summaries often include word-for-word excerpts from CNN articles. Because these summaries provide all the essential facts of a news story, users don’t need to click through to the original site. This behavior causes direct financial harm by diverting traffic and advertising revenue.
Evaluating Trademark Violation Allegations Against Generative AI Platforms
Regarding trademarks, the network claims that Perplexity’s interface frequently displays the CNN brand alongside AI-generated text. But this text may contain errors or hallucinations. CNN argues this unauthorized association risks tarnishing its reputation for accuracy. We see this as a strategic move to protect the “goodwill” associated with the CNN brand name.
Read More: CNIPA Deputy Commissioner Visits WIPO and EUIPO to Strengthen IP Cooperation
Best Practices for Maintaining Intellectual Property Compliance in AI Development
If your organization utilizes or develops AI tools, you must implement strict data governance policies to avoid similar liability. It’s not enough to assume that because information is publicly accessible on the web, it’s free for commercial exploitation.
- Review Your Data Sources: You should audit any automated data collection tools to ensure they respect “robots.txt” files and terms of service agreements. Ignoring these technical barriers can be used as evidence of willful infringement.
- Avoid Substantial Similarity: Don’t allow your platform to output results that mirror the structure or specific phrasing of a single source. Your tools should provide citations that lead users back to the original creator rather than replacing them.
- Protect Brand Integrity: You’re required to obtain express permission before using a third-party trademark in a way that suggests a partnership or endorsement. Ensure that AI-generated summaries are clearly labeled as such to prevent consumer confusion.
We expect this litigation to clarify the boundaries of “fair use” for the next decade. For now, the safest path is to secure licensing agreements before using protected content to power commercial products.