Meta Platforms is currently facing legal action from a number of prominent publishers who say that the company violated their copyright and used copyrighted materials to train its artificial intelligence model, known as Llama. The case was filed at the Manhattan federal court and will contribute to the ongoing legal battle regarding the learning process of artificial intelligence systems.
Among the plaintiffs are such popular publishers as Elsevier, Cengage, Hachette, Macmillan, and McGraw-Hill, along with author Scott Turow. These parties allege that Meta copied tens of millions of works illegally.
According to the complaint, Meta incorporated various types of content into its system. Specifically, the company allegedly used textbooks and academic journals, as well as literary fiction works. For instance, some of the materials mentioned in the suit are ‘The Fifth Season’ by N.K. Jemisin and ‘The Wild Robot’ by Peter Brown.
Not so, says Meta. One of its officials suggested that the training of AI models on existing content can be deemed as fair use under copyright law. The tech firm plans to vigorously defend itself in court. This strategy stems from a general approach by technology companies which view AI models as transforming output, not reproducing original materials.
AI’s Legal Reckoning and the Future of Copyright
As Maria Pallante, the president of the Association of American Publishers, pointed out, this debate is nothing like an innovation issue. According to Pallante, the use of copyrighted material without permission diminishes creativity because it deprives authors of income. Hence, AI will never flourish in such conditions.
Of course, this case is only one of the lawsuits brought against AI firms. The others are filed by writers, artists, and media outlets. They target companies like OpenAI, Anthropic, and several others. As in many other lawsuits of this kind, the legal issue concerns fair use of copyrighted material.
A consensus from the courts has not emerged yet. In this respect, two judges deliberated on issues similar to those at stake recently and came up with opposing conclusions, showing how unresolved this issue still is. The result of these cases will influence the further development of the artificial intelligence industry.
An interesting trend that should be discussed is the fact that the company named Anthropic decided to pay $1.5 billion to settle a class-action lawsuit initiated by authors. This agreement helped the company avoid lengthy negotiations in court that might cause even greater financial consequences for them. It might become one more example of such an approach.
Meta and the AI Legal Battle: Who Really Owns the Content?
For the publishing industry, the results of legal debates will be very important since it will determine whether this segment will be able to retain its right to control content or not. The outcome of the discussions on what is fair use will either strengthen or weaken this right.
Governments have been taking note as well because the topic is closely related to innovation, intellectual property rights, and economic development. Policy makers will have to balance these different concerns. Excessive control could limit the potential of AI. Too few regulations could be detrimental to creators and undermine the motivation for creating new content.
Moreover, corporations will come under scrutiny from stakeholders regarding their treatment of data and intellectual property rights. ESG principles will increasingly consider these issues. Lawsuits such as this one bring into the light the social component of ESG criteria by raising important ethical considerations.
This case will likely go through all the stages of the judicial process until it reaches its final outcome, which can take several years.
Regardless of the outcome, this particular case will have wider implications. It will clarify some aspects of how AI systems are trained and who gains from the use of these technologies.
