Almost everyone who reposts a viral clip believes they are safe because they changed something — trimmed it, flipped it, added a caption, kept it under fifteen seconds. After the Supreme Court’s 2023 decision in Warhol v. Goldsmith, that instinct is mostly wrong. Fair use for short video now turns on a single hard question, and it is not whether you edited the clip. It is whether your use does something genuinely different from the original — or just serves the same clip to the same audience for the same reason.
Fair use is not a percentage, a runtime, or a credit in the caption. It is a four-factor test written into 17 U.S.C. § 107 and applied case by case. The U.S. Copyright Office keeps a plain-language overview and index of decisions for exactly this reason: the same clip can be fair use in one context and infringement in another. This piece is a general map of how courts have drawn that line for short video. It is not legal advice, and reading it creates no client relationship.
The four factors are familiar: the purpose and character of the use, the nature of the original, the amount taken, and the effect on the market for the original. For viral video, two of them do most of the work — purpose (factor one) and market harm (factor four) — and after 2023 they increasingly point the same direction. Understand that, and most “is this fair use?” questions answer themselves before a lawyer is ever called.
The word that trips everyone up is “transformative.” In common usage it sounds like a description of editing: if I changed the file, I transformed it. Copyright law means something narrower and harder. A use is transformative when it has a different purpose or character — when it says something new about the original rather than simply repackaging it. The Supreme Court built that idea in Campbell v. Acuff-Rose Music (1994), which protected a rap parody precisely because parody comments on the work it borrows from.
The distinction is the whole game, and it is easy to miss. Mirroring a clip, trimming it, adding a filter, or stamping a logo changes how the clip looks. It does not change what the clip is for. The audience still watches it for the same reason the original audience did. None of those moves is the transformation that matters.
In Andy Warhol Foundation v. Goldsmith (2023), the Court considered Warhol’s silkscreen of a photograph of Prince, licensed to a magazine to illustrate a story about Prince. The Foundation argued the artwork added new meaning and so was transformative. The Court declined to treat “new meaning” as a trump card. The first factor, it held, asks whether the challenged use has a further purpose or different character — and where the secondary use shares substantially the same purpose as the original and is commercial, factor one points against fair use unless there is some other justification for the copying.
Read that against a viral repost. A clip that went viral did so because it entertains. An account that reuploads it does so to entertain the same kind of audience — and, on a monetized page, to do it for commercial benefit. Same purpose, commercial character. Under Warhol, that is the fact pattern courts have scrutinized most heavily, and the configuration in which fair-use defenses have tended to come up short. The decision did not invent a new rule so much as close a loophole: a stylistic tweak or a vague “new vibe” no longer carries the transformation argument on its own. The purpose has to actually differ.
Once purpose is the axis, the cases sort cleanly. On one side are uses courts have repeatedly rejected: reposting a clip whole, re-editing it to evade detection rather than to comment, stitching clips into a monetized compilation with no real analysis, or dropping someone’s footage into an ad. On the other side are uses with a genuine claim, because the clip became the subject of something new — criticism, parody aimed at the clip, real news reporting, scholarship.
Two video cases anchor the “genuine” side. In Hosseinzadeh v. Klein (S.D.N.Y. 2017), a reaction video that interrupted a sketch with point-by-point mockery was fair use, because the commentary was the product and no one would watch it instead of the original. In a 2015 California district-court decision (the Equals Three case), a show that joked about viral clips was found largely fair use — except for one segment where the host used a clip as a prop for unrelated jokes rather than commenting on the clip itself. That single losing segment is the most useful fact in the line: courts have rewarded commentary that is about the clip, not merely alongside it.
The right column is not a safe harbor. It is the zone where the answer depends on the details — how much was taken, how substantive the commentary is, whether the use competes with the original. A thorough critique with real analysis sits comfortably there; a full re-play with “so funny” tacked on the end is much weaker. The left column gathers the uses courts have most consistently rejected; the right gathers the ones where the outcome turns on the specifics.
Factor four asks whether the use steals demand from the original or its licensing market. For a viral clip, this is often decisive, because a repost on a bigger account does exactly that: it captures the views, the follows, and the ad share that would have flowed to the source. Courts call market substitution the core concern of copyright. A clip reuploaded for engagement looks like substitution close to its purest form — the same content, competing for the same audience. When factor one (same purpose, commercial) and factor four (direct substitution) point the same way, the remaining factors have seldom carried a use on their own.
Here is the part most enforcement programs miss: the doctrine that defeats most reposts also constrains how rights-holders may act. In Lenz v. Universal Music (9th Cir. 2016) — the “dancing baby” case — the court held that a copyright holder must consider whether a use is fair before sending a DMCA takedown, because fair use is authorized by law, not an excuse raised later. A notice sent without that consideration can expose the sender under Section 512(f).
That puts fair use at the front of any serious notice process, not at the end of it. The redline map above is one way to picture the triage that discipline calls for: the left side is where the case law points most clearly in one direction, and the right side is where a clip warrants a closer, case-by-case read — a human judgment, and sometimes a conversation rather than a notice. We have written before about why that kind of discipline is a way of building an asset rather than just stopping theft — and about what removal looks like in practice in our piece on platform removal times.
The principle underneath it is the one we keep returning to: it should pay more to partner than to infringe, and clearing rights should be the priority, with due process for everyone on both sides of a notice. A fair-use analysis is not a loophole for the infringer or a technicality for the rights-holder. It is the line that makes enforcement legitimate — and the line that tells a creator, before the views start climbing, whether the clip in front of them is theirs to use.
So the practical test is not “how much did I change it?” It is four plainer questions, drawn straight from the factors: Does my use say something about the clip, or just show it again? Did I take more than that point needs? Am I doing this for commercial gain? And does my version compete with the original for the same audience? Answer those honestly and the redline is usually obvious — which is the entire point of knowing where it sits before you cross it.
This article is general information about copyright law and the digital and social-media industry. Verights is the rights-enforcement brand of SocialCoaster Inc.; it is not a law firm and does not provide legal advice. Reading this article does not create an attorney-client relationship with Verights, SocialCoaster Inc., or any of its personnel, and nothing here describes, comments on, or takes a position regarding any specific party, dispute, or pending matter. Operational figures are aggregated and rounded to protect confidentiality; any illustrative figures are labeled as such. Forward-looking statements reflect current expectations and are not guarantees of any outcome; past results do not predict future outcomes. Consult qualified counsel about your specific situation.
Verights focuses on the rights behind viral video. Telling the clear-cut cases from the genuinely contestable ones is the work — and if that is the line you are trying to find, it is the conversation we have every day. We are the rights-enforcement brand of SocialCoaster Inc. This post is general education, not legal advice.
Talk to the Verights team