Copyright Issues For AI-Authored Journalistic Data Summaries.

28 Feb 2026 --
0 Comments

1. Copyright Protection for Facts vs. Expression

Journalistic summaries typically involve facts, which are not protected by copyright. Only the original expression or creative arrangement of facts is protected.

Feist Publications v. Rural Telephone Service

Facts:
A telephone directory publisher claimed copyright over its white pages listings.

Holding:
The U.S. Supreme Court held that facts are not copyrightable, and compilations require a “modicum of creativity” in selection or arrangement.

Application to AI Journalistic Summaries:
AI systems summarizing raw public data (e.g., election results, crime statistics) do not infringe merely by restating facts. However, copying a journalist’s unique narrative framing, analysis, or expressive language may infringe.

This case is foundational: journalistic data summaries are protectable only where there is original expression, not mere factual extraction.

2. Human Authorship Requirement in AI Outputs

If a news organization uses AI to automatically generate summaries, copyright depends on whether there is human creative involvement.

Burrow-Giles Lithographic Co. v. Sarony

Principle:
Copyright protects works reflecting human intellectual conception.

Relevance:
If an AI independently produces a news summary without meaningful human editing or creative input, it may lack copyright protection. However, if journalists design prompts, edit outputs, or structure narrative emphasis, those human contributions may satisfy authorship requirements.

Thaler v. Perlmutter

Holding:
Fully autonomous AI-generated works are not eligible for copyright registration because copyright requires a human author.

Impact on AI Journalism:
If a newsroom deploys AI to auto-generate summaries of stock prices or legislative proceedings without human editorial control, those outputs may enter the public domain in jurisdictions requiring human authorship.

3. Database and Compilation Rights

Journalistic summaries often rely on large datasets. The key question is whether copying or training on such databases infringes.

Football Dataco Ltd v. Yahoo! UK Ltd

Issue:
Whether football fixture lists qualified for database protection under EU law.

Holding:
The Court of Justice of the European Union held that database protection requires substantial investment in obtaining, verifying, or presenting contents—not merely creating the data.

Relevance to AI Summaries:
If AI systems extract structured data from protected databases, infringement may arise where substantial investment is appropriated. In Europe, the sui generis database right adds an extra layer of risk beyond copyright.

British Horseracing Board v. William Hill

Holding:
Database protection applies only where investment is in obtaining data, not in creating it.

Application:
If AI summarizes official government statistics (created, not “obtained”), database rights may not apply. However, scraping proprietary curated databases may infringe.

4. Fair Use and Transformative Summarization

AI journalistic summaries often rely on pre-existing articles. Whether such summarization is lawful may depend on fair use (U.S.) or fair dealing (Commonwealth jurisdictions).

Authors Guild v. Google, Inc.

Facts:
Google scanned millions of books to create a searchable database.

Holding:
The use was transformative and qualified as fair use.

Relevance:
AI systems that analyze articles to generate independent summaries may argue transformative purpose (information extraction rather than substitution). However, if summaries substitute for original journalism and reduce readership, fair use may fail.

Associated Press v. Meltwater U.S. Holdings, Inc.

Facts:
Meltwater distributed news excerpts and summaries to subscribers.

Holding:
The court found infringement; the summaries were not sufficiently transformative and harmed the market.

Application to AI News Summaries:
If AI-generated journalistic digests reproduce expressive portions or substitute for paid news subscriptions, liability may arise.

This case is especially relevant to automated news aggregation systems.

5. Substantial Similarity & Derivative Works

If AI summaries closely mirror the structure, tone, or distinctive phrasing of original articles, infringement may be found.

Harper & Row v. Nation Enterprises

Facts:
The Nation magazine published excerpts from President Ford’s unpublished memoir.

Holding:
Even small but qualitatively significant portions can infringe.

Application:
AI systems that extract key expressive segments from investigative journalism—even if short—may infringe if they capture the “heart” of the work.

Rogers v. Koons

Holding:
Substantial similarity in protected expression constitutes infringement, even if the format differs.

Relevance:
If AI summaries replicate creative narrative framing, distinctive metaphors, or investigative structuring, courts may find unlawful appropriation.

6. News of the Day Doctrine

Under copyright law, “news of the day” (facts about events) is not protected, but the reporter’s expressive treatment is.

International News Service v. Associated Press

Issue:
Whether copying time-sensitive news content constituted unfair competition.

Holding:
The Supreme Court recognized a quasi-property right in “hot news.”

Relevance:
AI systems generating real-time journalistic summaries may face liability under “hot news misappropriation” doctrines in limited jurisdictions if they free-ride on time-sensitive reporting investments.

7. Moral Rights & Attribution (India & Europe)

Where AI summaries rephrase investigative journalism, moral rights issues may arise.

Amarnath Sehgal v. Union of India

Holding:
Strong recognition of author’s moral rights, including integrity and attribution.

Application:
If AI systems distort investigative journalism or misattribute summaries, moral rights claims may arise in jurisdictions recognizing such protections.

8. Key Legal Issues in AI-Authored Journalistic Summaries

A. Originality

Mere factual restatement is not protected.

Expressive narrative structure is protected.

B. Human Authorship

Fully autonomous AI outputs may lack copyright.

Editorial oversight strengthens protection claims.

C. Database Protection

Particularly strong in the EU.

Extraction of substantial parts may infringe.

D. Fair Use / Fair Dealing

Transformative summarization may qualify.

Market substitution weighs heavily against fair use.

E. Hot News Misappropriation

Relevant for time-sensitive journalistic content.

Comparative Perspective

Jurisdiction	Key Position on AI Journalism
United States	Strict human authorship requirement; strong fair use doctrine
European Union	Database rights create additional protection layers
India	Human authorship required; moral rights strongly protected
UK	“Author” may be the person making arrangements, but interpretation remains debated

Conclusion

AI-authored journalistic data summaries exist in a legally sensitive zone:

Facts remain free for all.