Nine U.S. regional newspapers have filed a significant copyright lawsuit against OpenAI and Microsoft, seeking damages exceeding $10 billion. The complaint alleges that OpenAI’s models, including Microsoft’s Copilot, were trained on unlicensed news articles, with ChatGPT reproducing their content nearly verbatim. The lawsuit, brought by publications like the Boston Herald and Los Angeles Daily News, claims the AI memorized their articles, violating copyright laws and the Digital Millennium Copyright Act. Evidence includes side-by-side comparisons highlighting almost identical outputs.
Moreover, the lawsuit contends OpenAI and Microsoft ignored paywall agreements and systematically removed copyright information. The plaintiffs demand extensive damages and the destruction of GPT models containing their content. Additionally, OpenAI is embroiled in litigation regarding pirated book datasets allegedly sourced from Library Genesis. This unfolding battle raises critical questions about copyright in AI training and the protection of original content in the digital age.
Source link
