What Is the AI Copyright Lawsuit?
The AI copyright lawsuit landscape is a group of separate federal copyright cases against major AI companies, not one single lawsuit or class action. The largest centralized proceeding is MDL No. 3143, In re: OpenAI, Inc., Copyright Infringement Litigation, which consolidated twelve related suits against OpenAI and Microsoft before Judge Sidney H. Stein in the U.S. District Court for the Southern District of New York.
Separate from the OpenAI MDL, authors sued Anthropic over its Claude models, authors and visual artists sued Meta over its Llama models, and Getty Images sued Stability AI and, in a related case, artists and media companies sued Midjourney. Each proceeding has its own plaintiffs, its own court, and, so far, its own outcome; none of these resolves or controls the others.
“These actions share factual questions arising from allegations that OpenAI and Microsoft used copyrighted works, without consent or compensation, to train their large language models (LLMs), such as GPT-4, which underlie defendants' generative artificial intelligence products, such as OpenAI's ChatGPT and Microsoft's Bing Chat (rebranded as Copilot).”
Source: U.S. Judicial Panel on Multidistrict Litigation, In re: OpenAI, Inc., Copyright Infringement Litigation, MDL No. 3143, Transfer Order, Document 85 (April 3, 2025)
Why Are People Filing AI Copyright Lawsuits?
People are filing AI copyright lawsuits because they allege AI companies copied millions of copyrighted books, news articles, and images, without permission or payment, to train commercial AI models. Authors Steve Tremblay, Sarah Silverman, Michael Chabon, and the Authors Guild allege OpenAI trained its models on pirated copies of their books. The New York Times separately alleges OpenAI and Microsoft's products can reproduce verbatim or closely paraphrased excerpts of its articles, a claim about the models' output, not just their training data.
Authors brought a parallel claim against Anthropic, alleging the company built its training library in part from books downloaded from pirate sites such as Library Genesis, a claim a federal judge found largely true before the case settled. Visual artists sued Stability AI and Midjourney over image-generation models they allege were trained on copyrighted artwork without consent, and Getty Images separately sued Stability AI over its stock photo library.
The MDL's own transfer order notes these actions involve claims that defendants trained large language models on copyrighted text without consent or compensation, and that at least one group of plaintiffs, the News plaintiffs, also allege the models' own output reproduces their copyrighted content, a distinct theory from the training-data claim the book-author plaintiffs bring.
What Injuries Are Included in the AI Copyright Lawsuit?
The AI copyright lawsuit landscape includes three distinct legal claims, pled in different combinations across the various cases. Each is defined in plain language below, followed by what plaintiffs specifically allege.
Training-Data Copyright Infringement
Training-data copyright infringement is a legal claim that a company copied protected creative works, without a license, to build the dataset used to train an AI model.
Authors, news organizations, and visual artists allege OpenAI, Anthropic, Meta, Stability AI, and Midjourney each copied millions of copyrighted books, articles, or images into their AI models' training data without consent or compensation. A federal judge found in Anthropic's case that the company had downloaded and stored millions of books from pirate sites such as Library Genesis before that claim was resolved through the company's $1.5 billion settlement.
Output-Based Copyright Infringement
Output-based copyright infringement is a claim that an AI product's generated responses, not just its training process, reproduce substantial protected content from a copyrighted work.
The New York Times alleges that ChatGPT and Microsoft's Copilot can generate verbatim or closely paraphrased excerpts of Times articles in response to user prompts, a claim distinct from the training-data theory other plaintiffs bring. This theory is central to the cross-motions for summary judgment on fair use the parties filed in September 2026, since output-based copying is harder to defend as a transformative use than training alone.
DMCA Copyright-Management-Information Removal
A DMCA copyright-management-information claim alleges a company removed identifying information, such as an author's name or copyright notice, from a work before using it, a separate federal claim from copyright infringement itself.
Plaintiffs in the Raw Story Media and The Intercept actions, folded into the OpenAI MDL, do not allege copyright infringement at all. Instead, they allege OpenAI violated the Digital Millennium Copyright Act by stripping author, title, and copyright information from articles before using them, a narrower and legally distinct claim from the infringement theories pursued by the book-author and news-organization plaintiffs in the same MDL.
Who Qualifies for the AI Copyright Lawsuit?
You may qualify if:
- You are an author, journalist, visual artist, or other rights holder whose copyrighted, registered work was used, without your consent, to train a major commercial AI model
- You can identify specific AI products (such as ChatGPT, Claude, Llama, Stable Diffusion, or Midjourney) that plausibly drew on your work, based on public reporting or a dataset disclosure
- Your claim falls within the applicable copyright statute of limitations, generally three years from when you discovered or reasonably should have discovered the infringement
You may not qualify if:
- Your work was never formally registered with the U.S. Copyright Office, which federal law generally requires before filing an infringement suit, though registration can sometimes be completed shortly before filing
- You already received payment through the approved Anthropic settlement for the same specific works, which closed that claim for the works it covers
- You cannot point to any evidence connecting your specific work to a specific AI company's training data or model output
Eligibility depends on your work's copyright registration status, the specific AI company and model involved, and your jurisdiction's statute of limitations. Only an attorney experienced in AI copyright litigation, reviewing your specific work and its use, can confirm whether you qualify to file or join an existing case.
What Is the Current Status of the AI Copyright Lawsuit?
The current status of AI copyright litigation varies sharply by defendant and case. Anthropic's track has already settled and closed, while the OpenAI MDL, the New York Times' case, and the Getty Images cases against Stability AI remain active, each at a different procedural stage.
In re: OpenAI, Inc., Copyright Infringement Litigation, MDL No. 3143
The Judicial Panel on Multidistrict Litigation centralized twelve related OpenAI and Microsoft copyright suits into MDL No. 3143 on April 3, 2025, assigning the docket to Judge Sidney H. Stein in the Southern District of New York. The consolidated actions include the Authors Guild class action, the New York Times' case, and suits from the Daily News, the Center for Investigative Reporting, Raw Story Media, and The Intercept.
| Proceeding Type | Federal multidistrict litigation, 28 U.S.C. § 1407 |
| Court | U.S. District Court, Southern District of New York |
| Judge | Honorable Sidney H. Stein |
| Centralized | April 3, 2025 |
| Actions Consolidated | 12, including Authors Guild, New York Times, and Daily News suits |
Bartz v. Anthropic PBC (Settled)
A federal judge approved Anthropic's $1.5 billion settlement with a class of authors and publishers on July 20, 2026, the largest copyright settlement on record. The same judge, William Alsup of the Northern District of California, had separately ruled that training an AI model on copyrighted text counts as fair use, even while finding Anthropic had illegally downloaded and stored pirated books.
| Proceeding Type | Certified federal class action, settled |
| Court | U.S. District Court, Northern District of California |
| Judge | Honorable William Alsup |
| Settlement Amount | $1.5 billion, final approval July 20, 2026 |
| Covers | An estimated 500,000 works, at roughly $3,000 per work |
Kadrey v. Meta Platforms, Inc.
Judge Vince Chhabria of the Northern District of California granted Meta summary judgment on fair use for training its Llama models on June 25, 2025, but limited the ruling to the thirteen plaintiffs in that case and faulted them for not building a record of market harm. The order left open that AI training could be found unlawful in other cases with a better-developed record.
| Proceeding Type | Federal copyright suit, summary judgment granted on fair use |
| Court | U.S. District Court, Northern District of California |
| Judge | Honorable Vince Chhabria |
| Ruling Date | June 25, 2025 |
| Scope | Binds only the 13 named plaintiffs in this case |
Getty Images (US), Inc. v. Stability AI, Ltd.
Getty Images' U.S. case against Stability AI remains active, with Judge Trina L. Thompson letting trademark, false-designation, and unfair-competition claims proceed in April 2026. Getty's related UK case against Stability AI saw its secondary copyright claim rejected by the High Court in November 2025, a ruling Getty is appealing, while the court found limited trademark infringement from Getty watermarks appearing in some outputs.
| Proceeding Type | Separate U.S. and UK copyright and trademark suits |
| U.S. Court | U.S. District Court, Northern District of California |
| U.S. Judge | Honorable Trina L. Thompson |
| UK Status | Secondary copyright claim rejected, November 2025, under appeal |
| U.S. Status | Trademark and unfair-competition claims proceeding as of April 2026 |
Have Any AI Copyright Lawsuits Been Settled?
Yes. Anthropic agreed to pay $1.5 billion to settle its author class action, approved by a federal judge on July 20, 2026, working out to roughly $3,000 per covered work across an estimated 500,000 books.
No other defendant in this landscape has reached a comparable settlement as of this writing. The OpenAI MDL, the New York Times' case, and the Getty Images cases remain unresolved, so any eventual recovery in those cases depends on a court's ruling on fair use or a future settlement neither of which has happened yet.
Despite the record size of the Anthropic settlement, many authors and creators have not treated it as an unqualified win, since the same judge ruled that training an AI model on copyrighted text can itself be fair use, a finding widely seen as favorable to the AI industry going forward even as it required Anthropic to pay for the specific pirated copies it stored.
Latest AI Copyright Lawsuit Updates
AI copyright lawsuit updates below cover the landscape's major filings, rulings, and settlements across every defendant, not one single case's timeline.
AI Copyright Lawsuit Timeline
OpenAI and the New York Times File Cross-Motions for Summary Judgment
OpenAI, Microsoft, and the New York Times filed cross-motions for summary judgment on fair use in the Times' case on September 4, 2026, with the U.S. Department of Justice separately filing a brief supporting OpenAI's position that training on copyrighted text does not itself infringe copyright.
Judge Grants Final Approval to Anthropic's $1.5 Billion Settlement
Judge William Alsup gave final approval to Anthropic's $1.5 billion settlement with the author class, the largest copyright settlement on record, resolving the claim that Anthropic illegally downloaded and stored pirated books to train its Claude models.
New York Times Amends Its Complaint Against Microsoft
The New York Times sought to amend its complaint to allege Microsoft encouraged OpenAI to use the newspaper's articles without authorization, while dropping two claims of trademark dilution and contributory copyright infringement from the suit.
Getty's U.S. Trademark and Unfair-Competition Claims Proceed
Judge Trina L. Thompson ruled that Getty Images' trademark, false-designation, and unfair-competition claims against Stability AI could move forward in the U.S. case, even as the parallel UK case took a narrower path on copyright.
UK High Court Rejects Getty's Secondary Copyright Claim
The UK High Court rejected Getty Images' secondary copyright infringement claim against Stability AI, finding the Stable Diffusion models do not store reproductions of the training works, while finding limited trademark infringement from Getty watermarks appearing in some outputs; Getty is appealing.
Judge Grants Meta Summary Judgment on Fair Use in Kadrey Case
Judge Vince Chhabria ruled that Meta's use of books to train its Llama models was fair use for the thirteen plaintiffs in Kadrey v. Meta, while explicitly noting that training on copyrighted text could be found unlawful in other cases with a stronger record of market harm.
Judicial Panel Centralizes OpenAI Copyright Suits Into MDL No. 3143
The Judicial Panel on Multidistrict Litigation centralized twelve related copyright suits against OpenAI and Microsoft into MDL No. 3143, assigning the docket to Judge Sidney H. Stein in the Southern District of New York.
Judge Denies Most of OpenAI and Microsoft's Motion to Dismiss
Judge Sidney H. Stein denied most of OpenAI and Microsoft's motion to dismiss the New York Times' case, allowing the central copyright infringement claims to proceed while narrowing the case by dismissing several Digital Millennium Copyright Act claims.
New York Times Sues OpenAI and Microsoft
The New York Times filed suit against OpenAI and Microsoft, alleging the companies used millions of Times articles to train their AI models and that their products can reproduce the newspaper's content without authorization.
How to File an AI Copyright Lawsuit
Below are the 5 steps involved in filing an AI Copyright lawsuit.
- Confirm your work's copyright registration: Federal copyright law generally requires registering a work with the U.S. Copyright Office before filing an infringement suit, so confirm your registration status first.
- Identify the specific AI company and model involved: Pinpoint which AI product, such as ChatGPT, Claude, Llama, or an image generator, plausibly used your work, based on public dataset disclosures or reporting.
- Gather evidence connecting your work to the AI model: Save any output, documentation, or published research showing the model reproducing or closely mirroring your copyrighted work.
- Consult an attorney experienced in AI copyright litigation: An attorney can confirm whether your work and its use fit an existing case, such as the OpenAI MDL, or support a new, individual filing.
- File in the appropriate federal court: A qualifying new case against OpenAI or Microsoft would typically be filed and then considered for transfer into MDL No. 3143 for coordinated pretrial proceedings.
Answer a few questions and a case reviewer will follow up. Free, no obligation.
A case reviewer will follow up soon.
What Evidence Do I Need for an AI Copyright Lawsuit?
The strongest evidence in an AI copyright case is documentation connecting your specific copyrighted work to a specific AI company's training data or product output. That includes your work's copyright registration certificate, any dataset disclosure or research showing your work was used for training, and, for an output-based claim, saved examples of the AI product reproducing or closely paraphrasing your work.
What Have AI Companies Acknowledged in These Cases?
The companies' positions differ sharply by case. Anthropic did not admit liability in its settlement, but the presiding judge had already found the company downloaded and stored millions of pirated books before the case resolved. Meta won a fair-use ruling on training, but the judge limited that ruling to the specific plaintiffs and record in that case, not a blanket finding that AI training can never infringe copyright.
OpenAI and Microsoft continue to dispute the New York Times' claims, and the U.S. Department of Justice filed a brief in September 2026 supporting OpenAI's position that training on copyrighted text generally does not constitute infringement. Stability AI has disputed Getty's claims in both the U.S. and UK cases, and the UK High Court's November 2025 ruling found in Stability AI's favor on the core secondary copyright claim while still finding limited trademark infringement.
| Defendant | Key Outcome So Far | Year |
|---|---|---|
| Anthropic | $1.5 billion settlement, approved | 2026 |
| Meta Platforms | Summary judgment for fair use (limited scope) | 2025 |
| Stability AI (UK) | Secondary copyright claim rejected; limited trademark finding | 2025 |
| OpenAI / Microsoft | Motion to dismiss mostly denied; case active | 2025-2026 |
How Do the OpenAI, Anthropic, Meta, and Getty Cases Compare?
These are four separate, legally distinct proceedings with different defendants, different courts, and different outcomes so far, and should not be treated as one case or confused with each other.
The OpenAI MDL (this page's primary proceeding) consolidates book-author and news-organization claims against OpenAI and Microsoft in New York and remains active. Anthropic's case, brought by a separate group of authors in California, has already settled for $1.5 billion. Meta's case, also in California, produced a narrow fair-use win limited to its own thirteen plaintiffs. Getty's cases against Stability AI, split between the U.S. and UK, remain active with mixed early results.
| Case | Defendant | Court | Status on This Page |
|---|---|---|---|
| In re: OpenAI Copyright Litigation, MDL 3143 (this page's main case) | OpenAI, Microsoft | S.D.N.Y. | Active, pending summary judgment |
| Bartz v. Anthropic PBC (related, distinct) | Anthropic | N.D. Cal. | Settled, $1.5 billion, approved |
| Kadrey v. Meta Platforms (related, distinct) | Meta Platforms | N.D. Cal. | Partial summary judgment for Meta |
| Getty Images v. Stability AI (related, distinct) | Stability AI | N.D. Cal. and UK High Court | Active in U.S.; mixed ruling, under appeal in UK |
Frequently Asked Questions
What is the AI copyright lawsuit about?
The AI copyright lawsuit landscape covers several separate federal cases alleging major AI companies, including OpenAI, Anthropic, Meta, and Stability AI, copied copyrighted books, news articles, or images without consent or payment to train their AI models.
Is the AI copyright lawsuit a class action?
Some of it is. Anthropic's case was a certified class action that settled for $1.5 billion, and the Authors Guild case within the OpenAI MDL is also a class action, while other proceedings, such as Kadrey v. Meta, involve a smaller, named group of plaintiffs rather than a certified class.
Has the AI copyright lawsuit been settled?
Only Anthropic's track has settled, for $1.5 billion with final court approval in July 2026. The OpenAI MDL, the New York Times' case against OpenAI and Microsoft, and the Getty Images cases against Stability AI remain active and unresolved.
Who are the defendants in the AI copyright lawsuits?
OpenAI, Microsoft, Anthropic, Meta Platforms, Stability AI, and Midjourney are among the major named defendants, with authors, news organizations such as the New York Times, and visual-content companies such as Getty Images as the recurring plaintiffs.
Did Meta win its AI copyright lawsuit?
Meta won summary judgment on fair use in Kadrey v. Meta in June 2025, but the ruling applies only to the thirteen named plaintiffs in that specific case, and the judge noted that AI training could still be found unlawful in other cases built on a stronger record of market harm.
Can I still join an AI copyright lawsuit in 2026?
Possibly, depending on your specific work and which AI company used it. The OpenAI MDL and the Getty Images cases remain open and active, while Anthropic's settlement has already closed that specific claim for the works it covers, so an attorney needs to review your situation individually.