Looking for advice on this situation. Company Scraped My Blog Posts to Train Their AI - Copyright Violation? Any guidance would be greatly appreciated.
Details: I'm in a situation where I need to understand my legal options. Has anyone dealt with something similar?
Fair use is one of the most misunderstood concepts in IP law. It's a defense, not a right. You can't know for certain whether something is fair use until a court rules on it. The four-factor test is inherently case-specific.
Registration with the Copyright Office is important for enforcement. You can't sue for statutory damages or attorney fees without registration. The filing fee is $65 and it's worth doing for any valuable creative work.
This is a common issue that comes down to whether the work was "work for hire" or independent contractor work. The Copyright Act has specific definitions for each, and the distinction matters enormously.
Fair use is one of the most misunderstood concepts in IP law. It's a defense, not a right. You can't know for certain whether something is fair use until a court rules on it. The four-factor test is inherently case-specific.
Fair use is one of the most misunderstood concepts in IP law. It's a defense, not a right. You can't know for certain whether something is fair use until a court rules on it. The four-factor test is inherently case-specific.
Registration with the Copyright Office is important for enforcement. You can't sue for statutory damages or attorney fees without registration. The filing fee is $65 and it's worth doing for any valuable creative work.
Registration with the Copyright Office is important for enforcement. You can't sue for statutory damages or attorney fees without registration. The filing fee is $65 and it's worth doing for any valuable creative work.
This is a common issue that comes down to whether the work was "work for hire" or independent contractor work. The Copyright Act has specific definitions for each, and the distinction matters enormously.
This is a common issue that comes down to whether the work was "work for hire" or independent contractor work. The Copyright Act has specific definitions for each, and the distinction matters enormously.
Update: Thanks everyone for the guidance. I consulted with an attorney and we're moving forward. The advice here helped me understand what questions to ask and what to expect. Will update when there's a resolution.
Just discovered via server logs that an AI company's bot has been scraping my entire website (5,000+ articles) despite my robots.txt blocking their user agent. They're ignoring it completely. I have copyright on all content and a TOS that prohibits automated scraping.
What are my actual legal options here? The NYT v. OpenAI case is still ongoing, so is there any precedent I can rely on?
You have several potential claims, each with different strengths:
1. Copyright infringement: The strongest claim. Copying your articles into a training dataset is reproduction under 17 U.S.C. § 106. The AI company will argue fair use (transformative purpose, no market substitution). Post-NYT v. OpenAI, courts are still sorting this out, but publishers with registered copyrights have the strongest position.
2. CFAA (Computer Fraud and Abuse Act): If they ignored your robots.txt AND your TOS prohibits scraping, accessing your server after being notified of restrictions may violate the CFAA. The hiQ v. LinkedIn decision muddied this for public data, but your TOS + robots.txt combination is stronger than LinkedIn's case.
3. Trespass to chattels: If the scraping imposed a meaningful burden on your servers (increased costs, slowdowns), this common law claim applies.
4. State unfair competition: Building a commercial product using your content without compensation may constitute unfair business practices.
Immediate steps: (1) Send a formal cease and desist, (2) implement technical blocking (IP blocks, Cloudflare bot protection), (3) register your most valuable content with the Copyright Office if not already done.
Has anyone tried mediation before going to court? In my county, the court actually requires mediation for disputes under $25K. It was faster and cheaper than I expected, and we reached a resolution in one 3-hour session.
Journalist here with practical experience on this issue. My outlet discovered that several AI companies had scraped our entire archive of investigative articles going back to 2015. We documented this by using specific prompts designed to elicit verbatim or near-verbatim reproduction of our content, then compared the outputs against our published articles.
Our legal team pursued two parallel tracks. First, we sent DMCA takedown notices to companies where we could identify specific outputs that reproduced our content. Second, we joined a coordinated legal action with other news organizations challenging the use of our content for AI training without licensing.
The most important development is the concept of output liability. Even if courts decide that ingesting copyrighted content for training purposes might be fair use, there is a separate argument that generating outputs that compete with or substitute for the original content constitutes infringement. This theory has the most traction in current litigation.
For individual content creators: document everything. Use the Wayback Machine to establish publication dates. Register your most valuable content with the Copyright Office, since 17 U.S.C. Section 412 requires registration before you can claim statutory damages. The filing fee is $65 per work, and you can file groups of related works together.