Reddit sues Anthropic for scraping website content without consent

What just happened? It's not just OpenAI that keeps being sued for allegedly scraping content to train its systems. Reddit is suing Anthropic over claims it scraped content generated by forum users to train its Claude chatbot without consent. In a complaint filed in San Francisco this week, Reddit claims that Anthropic intentionally trained its LLMs on content created by Reddit users without requesting consent, thereby violating Reddit's user agreement. The filing adds that Anthropic accessed Reddit more than 100,000 times, despite the AI firm stating that its bots had been blocked from scraping data from the site. It adds that Anthropic has been training its Claude chatbot on Reddit data since at least December 2021. The suit actually includes a screenshot appearing to show Claude acknowledging that it was trained on Reddit user data. Reddit signed agreements with Google and OpenAI in 2024, worth $60 million and $70 million per year respectively, that allow them access to posts cre ... Read full article.