Highlights
Reddit has initiated legal action against AI startup Anthropic, alleging that the firm has illicitly accessed its platform over 100,000 times since July 2024 to collect user-generated content. This comes despite previous assurances that its bots had been blocked.
The lawsuit, submitted on Wednesday in San Francisco Superior Court, describes Anthropic as a “late-blooming” artificial intelligence company that publicly champions ethical practices while privately sidelining them for profit.
Reddit’s Claims Regarding Anthropic’s Conduct
The filing states, “This case highlights the dual nature of Anthropic: the public persona that aims to win consumer trust through claims of integrity and legal adherence, and the internal reality that disregards any regulations that hinder its profit motives.”
As of the time of writing, Anthropic has not commented on the lawsuit.
Value of Reddit’s Content in AI Development
Ben Lee, Reddit’s chief legal officer, conveyed to The Verge that the unauthorised exploitation of Reddit’s extensive repository of discussions could result in “billions of dollars” in commercial worth. “Reddit’s humanity is critically important in an era dominated by AI,” Lee mentioned, emphasising that “more than ever, individuals desire genuine human-to-human interaction. Reddit contains nearly 20 years of diverse, authentic discussions on nearly every conceivable topic. Such dialogues are unique to Reddit and vital for training language models like Claude.”
A Brief Overview of Anthropic
Anthropic, with backing from Amazon, is the creator of Claude, a conversational AI assistant that competes with OpenAI’s ChatGPT. The company is already facing other legal challenges from both publishers and authors. In August 2023, a group of writers filed a lawsuit claiming that Anthropic’s AI systems were trained using copyrighted literature. Additionally, in October of the previous year, Universal Music initiated legal action, accusing the company of using copyrighted song lyrics without consent.
Reddit’s Monetisation Efforts
This legal dispute follows Reddit’s own initiatives to monetise its data. In February 2024, Reddit entered into a partnership with Google, reportedly valued at approximately £60 million annually, to provide access to its content for AI training.
The Reddit versus Anthropic case is part of an expanding collection of legal disputes between content creators and AI organisations regarding copyright and data usage. Companies such as OpenAI, Meta, and Cohere have also faced lawsuits from media firms, book publishers, and authors striving to clarify the limits of data utilisation for training expansive language models.
