OpenAI and Anthropic Face Rising Legal and IP Challenges Amid AI Growth
OpenAI and Anthropic confront major copyright settlements, policy reversals, and insurer hesitancy amid expanding AI legal risks.
- • Anthropic settles for $1.5B over piracy claims involving 500,000 books, compensating eligible authors $3,000 each.
- • OpenAI reverses Sora 2's initial IP opt-out policy after backlash from Hollywood studios over unauthorized character use.
- • OpenAI CEO Sam Altman proposes more control and revenue sharing with rights holders to address IP concerns.
- • Both AI firms struggle with multibillion-dollar lawsuits and face insurer reluctance to provide comprehensive AI liability coverage.
Key details
OpenAI and Anthropic are currently grappling with significant intellectual property (IP) and legal challenges as the rapid advancement of AI technologies raises concerns about copyright infringement and liability risks.
Anthropic recently agreed to a $1.5 billion settlement after being sued for illegally downloading roughly 500,000 copyrighted books from piracy sites Library Genesis (LibGen) and Pirate Library Mirror (PiLiMi) to train its AI chatbot Claude. The federal judge approved the settlement on September 25, allowing affected authors to file claims to receive $3,000 per included book. However, only authors meeting strict eligibility criteria, including US Copyright Office registration and publication timeframes, can claim compensation. They have until March 23, 2026, to submit claims using unique IDs in official notices (ID: 92105).
Meanwhile, OpenAI's new text-to-video AI, Sora 2, aimed at generating viral and humorous videos, raised alarms in Hollywood due to its use of well-known characters from franchises like Star Wars and Pokémon without proper authorization. Initially, OpenAI implemented an opt-out system for IP owners, but the approach was criticized for being ineffective and burdensome, prompting CEO Sam Altman to reverse the policy. Altman acknowledged the need to provide rights holders with more granular control over their content and suggested potential revenue-sharing models with studios like Disney and Warner Bros., both of which have flagged unauthorized use and claimed copyright infringement (IDs: 92080, 92082).
In addition to copyright disputes, both OpenAI and Anthropic face potential multibillion-dollar lawsuits related to AI risks. The companies are exploring the use of investor funds to cover these liabilities, as insurance providers have expressed reluctance to offer comprehensive coverage for AI-related legal exposures, reflecting the uncertain and growing risks in this sector (ID: 92103).
These developments highlight the intensifying tensions between AI innovation and intellectual property rights, with major AI firms adapting policies and seeking financial solutions to manage unprecedented legal and ethical challenges posed by AI-generated content.