Fake Facts and Copyright Traps

In copyright, you cannot own a fact. In our earlier classes we learned this in Maltz v Witterick,¹ where the Federal Court held that “the use of common historical facts is not copyright infringement.”²

Winkler v Hendley³ builds on this, and further holds that an author who publishes what is said to be a nonfiction historical account cannot later claim the account is actually fictional to avoid the rule that there is no copyright in facts. In essence, if something is held out to be a fact, it is treated as a fact for the purposes of copyright whether it is true or not. 

This whole discussion unlocked an old memory of this classic Vsauce video about words and language, particularly the part about copyright traps. Dictionaries including fake words, mapmakers including fake features (paper streets, paper towns), all in an attempt to catch people copying their work.

Most relevant here, the author of The Trivia Encyclopedia, Fred Worth, included a fake fact because he was convinced the Trivial Pursuit board game was copying from his books. Worth wrote that TV detective Columbo’s first name was Philip, when in reality his first name was never mentioned in the show (at least not canonically). Sure enough, the following showed up in the next Trivial Pursuit game:

Worth sued for $300 Million for copyright infringement. Expectedly, the claim failed, and the 9th Circuit Court of Appeals upheld the ruling.⁴ As Michael aptly says in the video, you cannot own a fact, and you cannot own a lie you made up if everyone believes it. Despite failing in the courts, this story and Worth’s fake fact regarding Columbo’s first name lives on in the minds of many Columbophiles.⁵

Seems all this is may be a live issue again today, with more and more people relying on AI as a sort of search engine to find facts and information. One example we saw was Encyclopedia Britannica suing Perplexity AI alleging it misused their content in its answer engine for internet searches. Notwithstanding the fair use/dealing issue, this line of precedent seems to suggest that a copyright infringement would not even be found in the case of stolen facts, even if Britannica tries sneaking in a few copyright traps.

This also makes me wonder about the prevalence of “fake facts” today, particularly resulting from AI hallucinations and the tendency for LLMs to present false information with utmost confidence. There is a public interest in allowing people to rely on the veracity of factual accounts without needing to independently confirm them, lest they be liable in copyright.⁶ How this might play out in the world of AI hallucinations and misinformation is not certain, but I imagine the courts will uphold the old rule: there is no copyright in facts, true facts and fake facts alike.

– Dylan (Images of Columbo and his fake name tag generated by ChatGPT, text written by me)

1. 2016 FC 524 [Maltz]. ↩︎
2. Maltz at para 47. ↩︎
3. 2021 FC 498 [Winkler]. ↩︎
4. Worth v. Selchow & Righter Company, 827 F.2d 596 (9th Cir. 1987). ↩︎
5. see, e.g., this Reddit post from which I got the images of the Trivial Pursuit card, and this Columbophile Blog post. ↩︎
6. Winkler at para 95. ↩︎