Publishers vs. Cohere: Legal Battle Over Use of Generative AI | Google generative ai certification | Generative ai benefits for business examples | Generative ai course free with certificate | Turtles AI
A group of 14 publishers has filed a lawsuit against AI startup Cohere, accusing it of copyright infringement on thousands of works. The company is defending itself, calling the lawsuit baseless.
Key Points:
- Copyright Infringement Claim: Publishers allege that Cohere used more than 4,000 protected articles to train its models.
- Traffic Damage to Publishers: Plaintiffs allege that Cohere replicates significant portions of content, reducing the number of visitors to their sites.
- Trademark Challenge: The AI company is accused of generating content that is falsely attributed to the news organizations involved.
- Cohere’s Response: The startup denies the allegations, defending its AI training methodology as copyright-friendly.
A consortium of 14 major publishers, including Condé Nast, The Atlantic, and Forbes, has filed a lawsuit against generative AI technology company Cohere. The main complaint alleges the unauthorized use of at least 4,000 copyrighted articles to train the startup’s language models. According to the plaintiffs, Cohere not only used this content without permission, but also provided users with large portions of the original articles, and in some cases entire texts, compromising referral traffic to the publishers’ sites and threatening their business model. In addition to copyright infringement, the complaint highlights another problem: the generation of “hallucinated” content, or content falsely attributed to the news organizations, potentially impacting their reputation and trustworthiness.
Cohere’s response was swift. Josh Gartner, the company’s communications director, said the company firmly believes its approach is correct and considers the lawsuit “misguided and baseless.” He also noted that Cohere has always taken steps to reduce the risk of intellectual property infringement and would have preferred to address publishers’ concerns through direct dialogue rather than learning about them through a court case.
This legal dispute is part of a broader dispute between AI companies and content rights holders. Some industry players, such as OpenAI, have opted for licensing agreements with publishers to prevent conflict and consolidate their position, while others continue to defend their actions by invoking the principle of fair use.
Regardless of the outcome of the Cohere case, the debate over intellectual property in the AI era remains heated and destined to redefine the relationship between technology and information.