Your Words Have Value.
Let's Protect Them From AI.
OpenAI, Anthropic, Meta, Google, and others scraped billions of pages of copyrighted content: articles, books, song lyrics, newsletters, blog posts. They used this content to build products worth hundreds of billions of dollars. Federal courts are now hearing these cases. Creators with registered copyrights may be entitled to recover under 17 U.S.C. §504.
You don't need to prove financial loss.
Under 17 U.S.C. §504, copyright holders are entitled to statutory damages, a fixed range set by Congress, without having to calculate or prove what they actually lost. This is the core reason these cases have meaningful value.
The statutory ceiling under 17 U.S.C. §504(c)(2) for willful copyright infringement. Courts have discretion within the range. The more egregious the conduct, the higher the award.
The reported per-work figure in Anthropic's early author settlement. A single creator with 10 registered works could represent $30,000 in recovery at this rate.
Contingency representation only. Counsel earns nothing unless you do. Copyright registration through Copymight is also provided at no cost to qualifying claimants.
Past settlements do not predict future outcomes. Statutory damages are subject to court discretion. Consult an attorney about the value of your specific claim before making decisions based on these figures.
CopymightFree copyright registration for creatorsMost working creators qualify.
If you've published original content online, you may have a claim. A registered copyright is not required to get started. Copyright registration, if you don't already have it, is handled for you at no cost through our partner Copymight before your case proceeds.
Bloggers & Website Owners
Original articles, columns, or web copy published under your name or brand
Newsletter Writers
Substack, Mailchimp, Beehiiv, or any platform where you publish original writing to subscribers
Authors & Publishers
Published books, ebooks, short stories, or long-form works, whether traditionally or independently published
Songwriters & Lyricists
Original lyrics, compositions, or works where you hold the publishing rights
Journalists & Reporters
Articles published in news outlets, magazines, or editorial publications where you retain rights
YouTube Creators & Podcasters
Scripts, transcripts, or show notes you wrote and published alongside your content
Content Companies & Publishers
Organizations that own the rights to content produced by their writers or staff
Don't have a registered copyright? That's common and it's not a barrier. Copymight audits your published work and files group copyright registrations on your behalf, free of charge. Registration is what unlocks statutory damages under federal law.
Your copyright gets filed. At no cost to you.
Most creators never register their work with the US Copyright Office. That's understandable — it's a process most people don't know how to navigate. But registration is the legal prerequisite for statutory damages. Without it, your claim has significantly less value.
Copymight is the only organization operating at coalition scale to solve this problem. They audit your published content, file group copyright registrations on your behalf, and do it entirely free of charge.
Content audit
Copymight reviews your published works — articles, posts, lyrics, newsletters, books — and identifies what can be registered.
Group registration
They file with the US Copyright Office on your behalf. No forms to fill out, no fees to pay. Most registrations complete within a few weeks.
Your claim becomes eligible
Once registered, your works qualify for statutory damages of up to $150,000 per work under 17 U.S.C. §504(c)(2).
The companies being sued. The cases already filed.
The following AI companies are named defendants in active federal copyright litigation. These are not threatened or potential suits. They are cases currently moving through federal court, with discovery underway in several.
OpenAI
A consolidated class action brought by fiction authors alleging their books were used without license to train GPT models. Discovery is ongoing. A motion to dismiss was largely denied in 2024.
Anthropic
Music publishers sued Anthropic for reproducing copyrighted song lyrics through Claude. Anthropic reached an early settlement with a group of authors, paying approximately $3,000 per registered work.
Meta
Authors allege Meta used LibGen, a repository of millions of pirated books, to train its Llama models. A court order required Meta to produce internal training records in 2025.
Publishers and authors have filed suit alleging Google used copyrighted news content and YouTube transcripts without authorization to train Gemini and related AI systems.
Stability AI
Visual artists allege their work was scraped and used to train image generation models. The case has survived multiple dismissal attempts and is proceeding to discovery.
Perplexity AI
Several major publishers allege Perplexity reproduces full text of copyrighted articles in its AI-generated summaries without license, attribution, or compensation.
What's happened so far.
A summary of significant court rulings, settlements, and discoveries in active AI copyright cases. Updated as developments occur.
Anthropic settles author copyright claims: approximately $3,000 per registered work
Anthropic reached a confidential settlement with a group of plaintiff authors in Concord Music Group v. Anthropic. Reported per-work payments of approximately $3,000 for registered copyrights. This is the first major AI training data settlement to establish a per-work recovery figure, and is being cited as a benchmark in related pending litigation against OpenAI and Meta.
Federal court largely denies OpenAI motion to dismiss in Authors Guild case
Judge Colleen McMahon denied OpenAI's motion to dismiss claims related to verbatim output reproduction in Authors Guild v. OpenAI, No. 1:23-cv-08292. The court found plaintiffs adequately pled direct infringement. Discovery is now underway, including demands for OpenAI's training dataset documentation.
Meta ordered to produce LibGen training records in N.D. Cal. Llama litigation
In Kadrey v. Meta Platforms, No. 3:23-cv-03417, a federal magistrate ordered Meta to produce internal documents related to its use of LibGen, Z-Library, and other sources in training its Llama model family. Meta had argued the records were proprietary; the court disagreed. Production is ongoing.
Music publishers file suit over AI lyrics reproduction. Concord and UMG among plaintiffs.
Major music publishers including Concord Music Group and Universal Music Group filed suit in the Middle District of Tennessee alleging AI platforms reproduce copyrighted song lyrics verbatim in response to user queries. The suits target both training data use and real-time output reproduction as separate theories of infringement.
How this works, plainly.
Three things happen after you submit. None of them cost you anything.
Your content gets registered
If your published works aren't already registered with the U.S. Copyright Office, Copymight files group registrations on your behalf. This is a prerequisite to recovering statutory damages and is handled at no cost to you. Most registrations complete within a few weeks.
The Coalition reviews your file
The Coalition reviews your submitted works and copyright registrations — finding and recording where your content was used, handling any copyright filings or updates, and preparing your case. You won't deal with attorneys directly; the Coalition manages the entire legal process on your behalf.
You're notified of any recovery
If your case proceeds to settlement or judgment, you'll be notified and your share of recovery will be calculated based on your registered works. You pay nothing unless you recover. Attorneys are compensated from the recovery amount only.
Things people ask before submitting.
I'm not a published author. Can I still have a claim?
Yes. The word "author" in copyright law means the creator of any original work, not just book authors. If you write a blog, run a newsletter, post original lyrics, or publish anything original online, you are a copyright holder. The law covers all of it.
My work isn't registered. Does that mean I can't participate?
No. Unregistered work can still form the basis of a copyright claim, but registration is required to access statutory damages. That's why we handle registration through Copymight first. It's free, and it's what makes the recovery figures meaningful.
How do I know if my content was actually used to train an AI?
Honestly, you probably can't confirm it on your own, and you don't need to at this stage. Court-ordered discovery in active cases has already confirmed that billions of web pages, entire book libraries, and news archives were ingested without consent. If you've published original content online, the statistical likelihood is high. That question gets answered through litigation, not before you file.
What does it actually cost me to participate?
Nothing. Ever. Copyright registration is free through Copymight, and the attorneys work on contingency — they only get paid if you do. By joining, you're becoming a member of the Coalition, which pursues copyright infringement claims and legitimate AI licensing opportunities on behalf of its members.
What's the realistic recovery here?
The Anthropic settlement paid roughly $3,000 per registered work. That's an early benchmark, not a ceiling. Someone with 20 registered articles, blog posts, or songs at that rate would be looking at $60,000. Statutory law allows up to $150,000 per work for willful infringement, but courts set the actual number. We can't promise you a specific outcome.
What is Copymight and what do they do?
Copymight is our copyright registration partner at copymight.org, and part of the broader Coalition that pursues these claims. Copymight audits creators' published work and files group copyright registrations with the US Copyright Office at no cost to the creator — registration is what unlocks statutory damages under federal copyright law. Registration is only the first step: as a member, you're agreeing that the Coalition will act as plaintiff and pursue infringement claims on your behalf, not simply register your copyrights. That combination of registration and litigation is what gives these claims real value.