DooDooLamb News
Encyclopedia Britannica Sues OpenAI Over AI Training Data. Is Grokipedia Next?
Brief published March 18, 2026 ยท Original source published March 16, 2026
Original reporting by Bruce Gil at gizmodo.com.
Automated brief. Verify important details at the original source.
Encyclopedia Britannica Discovers What Every Content Creator Learned in 2023: ChatGPT Doesn't Ask Permission
The company that spent centuries convincing people to buy 32 leather-bound volumes of human knowledge has just realized that OpenAI slurped up all that knowledge without writing a check. Encyclopedia Britannica and its dictionary-making subsidiary Merriam-Webster filed a lawsuit this week claiming ChatGPT is "cannibalizing" their traffic, which is corporate speak for "people stopped paying us $39.95 a month when they could just ask a chatbot instead."
This legal drama represents the collision of two completely different business models: one where you charge people to access information, and another where you give away information for free after training your AI on everyone else's paid content. It's like watching a taxi company sue Uber, except the taxi has been around since 1768 and the Uber driver is a large language model that read every book ever written during a particularly productive weekend.
Encyclopedia Britannica's complaint centers on the argument that OpenAI scraped their premium content to train ChatGPT, then turned around and offered that same information for free to anyone with an internet connection. The irony is almost too perfect: the company that democratized knowledge by making it available in books (instead of keeping it locked away in monasteries) is now upset that someone democratized it further by making it available through chat interfaces. They've essentially been out-encyclopediaed by a company that doesn't even pretend to employ human editors, fact-checkers, or anyone who knows the difference between a reliable source and a Reddit comment.
The lawsuit claims OpenAI violated copyright law by using Britannica's articles to train its models without permission or compensation. This raises fascinating questions about what exactly constitutes fair use in the age of artificial intelligence. When a human reads an encyclopedia article and later incorporates that knowledge into their own writing, we call it research. When an AI system reads millions of encyclopedia articles and incorporates that information into its responses, apparently we call it theft. The distinction matters because it could determine whether every AI company needs to start cutting licensing deals with publishers, or whether they can continue their current approach of "ask for forgiveness, never for permission, and definitely hire better lawyers than the other guy."
Merriam-Webster's inclusion in the lawsuit adds another layer of absurdity to the whole affair. The dictionary publisher is essentially arguing that OpenAI shouldn't be able to use their definitions of words without payment, which is like claiming ownership of the building blocks of human communication. It's particularly rich coming from a company whose entire business model was built on organizing words that existed long before Noah Webster decided to start charging people to look them up. Now they're demanding ChatGPT pay licensing fees for explaining what "sesquipedalian" means, apparently unaware that their biggest competitive threat isn't AI cannibalization but the fact that Google has been showing dictionary definitions in search results for over a decade.
The timing of this lawsuit is telling. Encyclopedia Britannica didn't seem too concerned about digital disruption when Wikipedia launched in 2001 and immediately started eating their lunch with free, crowd-sourced content. They weathered that storm by pivoting to educational institutions and maintaining their reputation for editorial rigor. But ChatGPT represents a different kind of threat entirely: it's not just free and comprehensive, it's also conversational, accessible, and doesn't require users to parse through seventeen different disambiguation pages to find out why their science teacher mentioned something about mitochondria being the powerhouse of the cell.
What makes this legal battle particularly fascinating is that both companies are essentially selling the same thing: answers to questions people have. The difference is that Britannica employs humans to research, write, and fact-check those answers, while OpenAI trains neural networks on text data that may or may not include accurate information. One approach is more expensive and arguably more reliable; the other is faster and free. The market has spoken pretty clearly about which approach it prefers, which explains why Britannica is now trying to get the courts involved instead of competing on features or price.
The "traffic cannibalization" argument deserves special attention because it reveals just how dramatically the information landscape has shifted. Britannica is arguing that people who would have subscribed to their service are instead getting their information from ChatGPT, which learned how to provide that information by reading Britannica's content. It's a compelling argument until you realize that most people who ask ChatGPT factual questions weren't going to pay for a Britannica subscription anyway. They were going to Google it, check Wikipedia, or just remain ignorant. The real tragedy isn't that Britannica is losing subscribers to AI; it's that we've apparently decided convenience matters more than accuracy, and "good enough" answers from a chatbot beat thoroughly researched information from subject matter experts.
This lawsuit will likely set important precedents for how AI companies can use copyrighted content in their training data. If Britannica wins, it could force OpenAI and other AI companies to negotiate licensing agreements with publishers, potentially making AI development significantly more expensive and legally complex. If they lose, it essentially gives AI companies free rein to train on any publicly available text, accelerating the development of models that can replace traditional information services.
The deeper question is whether this represents innovation naturally displacing outdated business models, or whether we're witnessing the systematic devaluation of expertise and editorial quality. Encyclopedia Britannica spent over two centuries building a reputation for accuracy and reliability. ChatGPT spent a few months reading the internet and decided it was ready to answer questions about everything from quantum mechanics to medieval history. One of these approaches is more scalable; the other is more trustworthy. The lawsuit essentially asks courts to decide whether that distinction still matters in an age where people want immediate answers more than they want correct ones.