agentage.org

agentage.org / Articles / Checklist

Reporting from Paywalled Sources: What Is Allowed Under EU Law

A practical checklist for distinguishing between reportable public facts and protected paywalled text under current EU and Hungarian regulations.

By Zoltán Varga2026-11-055 min read

Short answerA news agency may report the facts contained within a paywalled article, such as who did what and when, because facts are not protected by copyright. However, the agency cannot summarize the specific text, arguments, or unique phrasing found behind the paywall without explicit permission or a license. You must attribute the discovery of the fact to the original source while writing the summary entirely in your own words.

Public Facts Versus Protected Paywalled Text

Copyright law protects the expression of ideas, not the ideas or facts themselves. A paywall restricts access to the written article, but it does not create a legal monopoly over the events described within it. If a reporter reads a paywalled piece and learns that a company launched a new product, that event is a public fact available for anyone to report.

The protection extends to the specific arrangement of words, the narrative structure, and the unique analysis provided by the original author. You cannot copy paragraphs, rephrase entire sections, or replicate the distinctive voice of the paywalled content. The boundary lies between the raw data of the event and the creative labor used to present it.

News agencies often rely on the principle that facts are free for the taking. This allows multiple outlets to cover the same story without infringing on intellectual property rights. The challenge arises when the line between reporting a fact and reproducing the source material becomes blurred through excessive quoting or structural mimicry.

Legal frameworks like the Hungarian Copyright Act explicitly state that protection does not extend to facts or daily news. This ensures that information flows freely to the public while still rewarding creators for their original writing. Your goal is to extract the signal from the noise without stealing the medium.

  • Facts about events remain free to report regardless of paywalls.
  • The specific wording and narrative structure are protected.
  • Rephrasing entire sections constitutes infringement.
  • Daily news and basic facts fall outside copyright scope.

No Summaries of Locked Content

Summarizing a paywalled article often crosses the line into copyright infringement if it captures the essence of the original work. A summary that mirrors the source's logic, flow, or detailed arguments effectively substitutes for the original text. This violates the rights of the publisher who invested resources to produce the content.

The EU Digital Single Market Directive introduced neighboring rights for press publishers. These rights protect the investment made in producing news content against unauthorized digital reuse. While short extracts are permitted, a comprehensive summary of a locked article typically exceeds the allowable limit for free use.

Automated tools that scrape and summarize paywalled content face significant legal risks. The act of bypassing a paywall to access the full text for the purpose of summarization can also violate terms of service. Even if the text is technically accessible, using it to generate a substitute product is generally prohibited.

Your workflow must avoid creating a condensed version of the paid article. Instead, focus on the specific data points or announcements that the article reveals. Treat the paywalled text as a source of verification rather than a draft to be rewritten. The output must be distinct from the input.

  • Comprehensive summaries of paywalled articles are prohibited.
  • Mirroring the source's logic violates neighboring rights.
  • Bypassing paywalls for text extraction carries legal risk.
  • Use locked content only for verifying specific data points.

Attributing Paywalled Reporting

When you report a fact discovered in a paywalled source, clear attribution is mandatory. You must explicitly state that the information came from a specific publication or reporter. Phrases like 'according to reports in [Publication Name]' or '[Reporter Name] disclosed' establish the chain of custody for the information.

Attribution serves two purposes: it respects the original journalist's effort and it clarifies the provenance of the fact for your readers. It prevents the appearance of claiming exclusive discovery when the information was actually gated. This transparency builds trust with your audience and protects your agency from plagiarism accusations.

The rule requires distinguishing between what the source claimed and what you independently verified. If you cannot verify the fact through open channels, you must frame it as an unconfirmed report from that specific outlet. This nuance is critical when dealing with sensitive or controversial news items.

Short quotes are permissible if they are necessary to illustrate a point, but they must be enclosed in quotation marks with a direct link. The quote should be minimal, serving only to highlight the original phrasing. Long passages or multiple quotes in a single article still constitute a derivative work that requires licensing.

  • Always name the specific publication and reporter.
  • Distinguish between source claims and independent verification.
  • Keep direct quotes minimal and clearly marked.
  • Provide a hyperlink to the original paywalled article.

Finding the Primary Source Instead

The most reliable method for reporting is to locate the primary source document rather than relying on secondary coverage. Press releases, official court filings, government databases, and corporate announcements are usually free and provide the raw facts directly. This approach eliminates the need to navigate paywalls or worry about copyright issues.

Primary sources offer the highest level of accuracy and reduce the risk of misinterpretation. When a journalist summarizes a paywalled article, they may introduce errors or bias. Going straight to the original document ensures your reporting reflects the actual data. This practice aligns with the standards of professional journalism.

If a primary source is not available, seek out alternative open-access reports covering the same event. Cross-referencing multiple independent outlets helps build a complete picture without infringing on a single paywalled article. This multi-source synthesis is a core requirement for high-quality news aggregation.

In cases where the only available information is behind a paywall, you may report the existence of the report itself. You can state that a publication has covered the topic without detailing the specific contents of their article. This acknowledges the news cycle without violating the rights of the publisher.

  • Prioritize press releases and official documents over news articles.
  • Cross-reference multiple open-access sources for verification.
  • Report the existence of a story if details are paywalled.
  • Avoid relying on a single secondary source for facts.

Edge Cases in Digital Reporting

Text and data mining for research purposes operates under specific exceptions in EU law. These exceptions allow automated processing of copyrighted works if the user has lawful access. However, the results of such processing cannot be redistributed publicly if they reproduce the protected expression of the original work.

AI-generated summaries face strict labeling requirements under the AI Act. Content generated by artificial intelligence regarding matters of public interest must be clearly labeled as such. This applies even if the underlying facts are accurate. Human editorial oversight remains the preferred method to ensure compliance and accountability.

The distinction between a hyperlink and a reproduction is vital. Linking to a paywalled article is generally legal and encouraged. Providing a snippet or a summary that acts as a substitute for clicking the link is not. The value of the news industry relies on driving traffic to the original source.

Licensing agreements with news agencies like Reuters or MTI provide a legal pathway to reuse content. These contracts define exactly how much text can be used and under what conditions. Without such an agreement, you must rely on the fair use principles of fact reporting and short quotation.

  • Data mining exceptions do not permit public redistribution of text.
  • AI-generated content on public issues requires clear labeling.
  • Hyperlinks are legal; substitute summaries are not.
  • Licensing is required for extensive text reuse from agencies.

Key takeaways

  • Facts are free to report, but the specific wording of paywalled articles is protected.
  • Never summarize the full content of a locked article; focus only on verifiable data points.
  • Always attribute facts to the original source and distinguish between claims and verification.
  • Prioritize primary sources like press releases to avoid copyright complications.
  • AI-generated news must be labeled, and human oversight ensures legal compliance.

Related questions

Can I rewrite a paywalled article in my own words?

You can rewrite the facts, but you cannot rewrite the structure or arguments of the article. If your rewritten version follows the same logical path and covers the same details as the original, it likely infringes on copyright. Stick to reporting the core event and cite the source.

What happens if I summarize a short paywalled article?

Even short articles are protected by neighboring rights if they represent a significant investment. Summarizing them entirely, regardless of length, risks legal action from the publisher. You may quote a sentence or two with attribution, but a full summary requires a license.

Do AI tools change the rules for reporting paywalled content?

AI tools do not change copyright laws. Using an AI to scrape and summarize paywalled content creates the same legal liabilities as doing it manually. Additionally, the resulting content must be labeled as AI-generated if it concerns public interest matters.

AI-assisted: drafted with a language model from the author's research notes, then edited and checked before publication. Author: Zoltán Varga, AI visibility researcher and AI ecosystem architect, Budapest.

More on agentage.org