Social media platform Reddit filed a lawsuit against AI company Anthropic on June 4, alleging "unauthorized scraping of platform data to train its AI model Claude," violating Reddit's user agreement and engaging in "unlawful and unfair business practices".
This move stimulated Reddit's stock price to rise 6.63% that day, closing at $118.21.
Core Allegations and Data Access
According to the lawsuit documents, Anthropic attempted to access Reddit data over 100,000 times without authorization from July 2024. Reddit noted that such actions continued even after being requested to stop. Reddit's Chief Legal Officer Ben Lee emphasized in a statement:
"We believe in an open internet, but this does not equate to open commercial exploitation. We will not tolerate profit-seeking entities like Anthropic commercially exploiting Reddit content to earn billions of dollars without giving back to Reddit users or respecting their privacy."
The focus of this lawsuit is not on copyright infringement, but rather on allegations of breach of contract (user agreement), property rights violation, unjust enrichment, tortious interference, and unfair competition. Therefore, Reddit is seeking damages, demanding that Anthropic return profits obtained from using its data, and requesting a court injunction to prevent unauthorized use of platform content.
Anthropic's Response
Previously, Reddit had reached AI data licensing agreements with Anthropic's main competitors, such as OpenAI and Alphabet (Google's parent company). These agreements specifically stipulate that partners must respect user privacy, support content deletion requests, and provide "compensation" to Reddit for data access and infrastructure costs.
Regarding this case, an Anthropic spokesperson stated that they "disagree with Reddit's allegations and will actively prepare a legal defense."
Founded by former OpenAI researchers, Anthropic initially positioned itself as more focused on AI safety and responsibility compared to competitors. However, Reddit's lawsuit criticizes them for inconsistent words and actions, disregarding platform rules and user privacy.
Content is "About Money"
This lawsuit reflects content platforms' widespread resistance to AI companies using web crawlers to extract content. Previous examples include The New York Times' lawsuit against OpenAI and Microsoft, and similar legal actions by multiple music publishers against AI audio companies (including Anthropic).
Reddit, with 20 years of history and extensive topic discussions, is undoubtedly an important data source for AI model training. Scraping without prior consent is clearly aimed at gathering evidence to demand compensation later, representing content platforms' awareness of data's commercial value in the AI era.
