Merriam-Webster’s dictionary defines the verb plagiarism as “to steal and pass off (another’s ideas or words) as one’s own: to use (another’s work) without citing the source.” And that’s exactly what its parent company, Encyclopedia Britannica, claims the AI company’s confusion with its AI Answers engine did, according to a complaint filed Thursday in the U.S. District Court for the Southern District of New York.
AI companies like Confusion are no strangers to copyright infringement. Openai, Midjourney, and others are currently fighting these kinds of legal claims in court. (Disclosure: CNET parent company Ziff Davis filed a lawsuit against Openai in April, alleging that it infringes on Ziff Davis’ copyright in the training and operation of its AI systems.)
However, Encyclopedia Britannica’s complaint is a bit different. The publisher is primarily concerned with the results of the confusing AI responses that the engine churns out, rather than how the underlying technology (a large language model) was trained.
Don’t miss any of our unbiased tech content and lab-based reviews. Add CNET as a preferred Google source.
Confusion is also somewhat unique among AI companies. Its revenue-sharing program aims to give media companies like Encyclopedia Britannica a tiny slice of its profit pie. At the same time, the publisher argues that the existence of the program proves that the AI company knows there is a market for human-generated content. Rather than paying for access to this copyrighted content, Confusion is likely accepting it without permission or payment.
What’s also different about this complaint is that Encyclopedia Britannica specifically calls out the potential reputational damage it is suffering from confusing trademark infringement. When you ask a question, the response may include a link to the original sources, as well as the websites’ logos. However, AI isn’t always right; it can produce things in error, called hallucinations.
Encyclopedia Britannica doesn’t want its brand names or trademarks anywhere near misinformation, writing in the complaint that the content created by Aiatre “confuses and misleads users into believing (falsely) that the hallucinations are company-affiliated, sponsored, or endorsed.”
The publisher also claims that Matlexity’s web scraping bot allows it to verbatim regurgitate its content. Similar concerns exist in the New York Times’ lawsuit against Chatgpt-maker Openai.
Neither Encyclopedia Britannica nor Confusion immediately responded to a request for comment.
Copyright law offers key protections to publishers and creators, and it has become one of the most contentious legal issues in AI. Typically, if a company wants to use copyrighted material, it must get permission from the rights holder and license it. Tech companies are pushing for “fair use” exceptions, a legal concept that would give them permission to use protected content without having to negotiate with the creators.
Claude-Maker Anthropic and Meta recently won copyright lawsuits under the fair use exception. The judge in Anthropic said the AI company’s use of copyrighted material was “extraordinarily transformative.” But both judges warned that tech companies won’t always win future cases.
