AAP Files Amicus Brief in Meta ai Copyright Case

THE Association of American publishers laid A friendly memory On April 11, supporting the authors in their collective appeal against Meta for copyright violation linked to AI training.

The thesis maintains that the use by meta of the work protected by copyright to form its LLAMA AI model does not comply with the standards of fairness and contradicts the affirmations of the company according to which the license options for such content do not exist.

“Meta’s systematic copy and the coding of protected creative works, word by word, in a large language model, is not a fair transformer use by virtue of the law, but rather exceeds the legal objective of the doctrine and the previous judicial,” said Maria A. Pallante, President and Chief Executive Officer.

The trial, submitted in December 2023 in the northern district of California, includes the complainants Sarah Silverman, Richard Kadrey, Christopher Golden, Michael Chabon, Ta-Nehisi Coates and Junot Díaz. They allege that Meta appropriated millions of works protected by copyright without authorization or compensation, obtaining a large part of the equipment from pirate sites such as the Anna and Libgen archives. This is one of the 40 current prosecutions brought by publishing industry organizations, businesses, authors and others against AI companies.

According to the thesis, “some researchers believe that the AI ​​training license market should be estimated at 2.5 billion dollars now and is expected to reach $ 30 billion in a decade.” The current meta market assessment is estimated at $ 1.4 billion.

The AAP file directly gives META’s assertion that there is no viable license option for AI training equipment, while declaring that “the existence of an active market for AI training materials is indisputable”. The document refers to the existing license agreements between AI companies such as Openai, Microsoft and Amazon with various publishers, as shown in the graph below:

The AAP notes that Meta would have had discussions with publishers on the acquisition of authorized content before finally choosing to use hacked materials instead. Much of this is described in Alex Reisner reports for The Atlantic.

The brief indicates that Meta’s actions “undermine copyright incentives that Congress has promulgated” and damages the ability of authors and publishers to benefit from their creative investments.

The main strengths of the dissertation include:

• Meta, “a company evaluated at more than one dollars billion, asks this court to declare that it is free to appropriate and commercially exploit the content of the work protected by copyright on a massive scale without authorization or payment.”
• The memory maintains that “common sense dictates that the words of the authors themselves, not only” statistical information “about them, are stored in the model”.
• The AAP maintains that the meta has escaped “technological protections which are essential to an online market functional for works protected by copyright”, which is “manifestly in contradiction with the congress mandate in the adoption of DMCA”.
• The brief concludes that “the long -term potential of AI technology will only be carried out by preserving the marketable rights which allow the authors, publishers and AI developers to engage in mutually beneficial commercial transactions”.

The memoir also maintains that a conclusion of fair use in this case “not only would undermine the public interest in an achievable copyright regime, but will encourage and reward theft twice”. Other organizations depositing AMICUS Memoirs include Copyright Alliance, the International Association of Scientific, Technical and Medical Editors and Copyright Professors.

A judicial hearing on the summary judgment requests made by each party is set for May 1.

The FTC affair begins today

The file was carried out on Friday. Today, Meta will start to address an antitrust case presented by the Federal Trade American Commission, which alleges that the acquisitions of Meta Instagram in 2012 and WhatsApp in 2014 were part of a strategy to remove competition and establish a monopoly on the social media market. The FTC argues that these acquisitions have been designed to eliminate potential rivals, violating federal antitrust laws and seeks to oblige Meta to give up these platforms to restore competition.