Class Action Lawsuit Certified Against Anthropic for AI Training Data Piracy
A federal judge has certified a class action lawsuit against Anthropic for allegedly pirating millions of copyrighted books for AI training.
Key Points
- • Federal judge certifies a class action lawsuit against Anthropic over copyright infringement.
- • Anthropic allegedly pirated 5 million books from LibGen and 2 million from PiLiMi.
- • Judge Alsup likened Anthropic's actions to 'Napster-style downloading'.
- • Potential damages could total up to $150,000 per work, leading to massive liability.
A US federal judge has officially certified a class action lawsuit against Anthropic, the AI company responsible for the Claude AI platform, for allegedly pirating millions of copyrighted books to train its AI system. The ruling, delivered by Senior District Judge William Alsup, allows authors whose works were unlawfully downloaded from major pirate libraries to join the class action, raising significant concerns around copyright infringement.
The lawsuit centers around claims that Anthropic downloaded approximately 5 million copyrighted books from the library LibGen, and around 2 million from another site, PiLiMi. This massive acquisition has drawn comparisons to 'Napster-style downloading', referring to the infamous file-sharing service that faced legal battles over copyright issues. Judge Alsup's ruling noted that while the use of copyrighted materials in training AI models could potentially be regarded as 'transformative' fair use, the initial acquisition of these pirated works does not meet fair use protections.
The financial implications for Anthropic could be severe, with potential damages estimated at up to $150,000 per infringed work. Given the millions of books at stake, this could result in catastrophic liability for the company, significantly surpassing previous high-stakes copyright cases, such as Oracle's $8.8 billion lawsuit against Google. Judge Alsup emphasized that the potential size of damages should not preclude class certification, aiming to uphold federal class action rules.
This landmark case could set crucial precedents for the AI industry at large, particularly regarding the legal methods by which companies can source training data. The outcome may influence future practices in AI training and reinforce the importance of copyright compliance in the rapidly evolving field of artificial intelligence. As the case unfolds, it will be closely scrutinized by industry stakeholders and legal experts alike, given its potential to reshape the landscape of AI training methodologies and copyright law.