Copyright infringement or technological breakthrough: OpenAI and Meta* will be held accountable in court for their technologies
Leading writers rebelled – neural networks take away their bread.
Celebrity comedian and author Sarah Silverman, along with writers Christopher Golden and Richard Kadray, have filed a lawsuit in U.S. District Court against accusations in copyright infringement by filing a lawsuit against companies Open AI and Meta*.
According to the charges, the training data used in the ChatGPT (OpenAI) and LLaMA (Meta), were illegally obtained from websites shadow libraries such as Bibliotik, Library Genesis (LibGen), Z Library and others, where the data was offered for download via torrents. These data contained the works of the mentioned authors.
In a lawsuit against OpenAI, the authors presented evidence ChatGPT offers summary summaries of their books upon certain requests, which the authors claim violates their copyrights. The books “Bedwetter” by Silverman, “Ararat” by Golden and “Sandman Slim” by Kadrey were cited as examples.
Parallel lawsuit vs. Meta is based on the fact that the authors’ works were available in the training data that Meta used for its LLaMA AI model, presented in February. The plaintiffs allege that the datasets were obtained illegally.
In both lawsuits, the authors emphasize that they “did not consent to the use of their copyrighted works as teaching materials” for AI models. The lawsuits include six counts of allegations including various types of copyright infringement, negligence, unfair enrichment and unfair competition. The authors demand payment of compensation payments, return of profits, etc.
Authors’ lawyers say they’ve received a lot of complaints from writers and publishers concerned about ChatGPT’s ability to generate text similar to that found in copyrighted material.
Such lawsuits are becoming a real problem for OpenAI and other AI companies, and they raise discussion about the limits of copyright in the age of artificial intelligence. Experts expect many more such lawsuits in the coming years. OpenAI and Meta did not provide comments.
Less than a week has passed since the filing of the previous lawsuit against OpenAI. In early July, science fiction writers Paul Tremblay and Mona Awad filed a class action lawsuit against OpenAI in federal court, accusing the company of copyright and privacy violations. According to the authors, OpenAI illegally used their books while teaching ChatGPT without obtaining proper consent, acknowledgment of authorship or compensation.
In addition, Microsoft and OpenAI in June found themselves at the center of a legal scandal because of their artificial intelligence products, which, according to 16 anonymous plaintiffs, illegally used and distributed their personal data without proper notice or consent.
Last week, OpenAI representatives said that AI with greater than human intelligence could appear within a decade. And if he really ever appears – he will not necessarily be benevolent, so need to explore ways to control and limit such AI . To do this, OpenAI is forming a new team to develop ways to manage and control “superintelligent” artificial intelligence systems.
* The Meta company and its products (Instagram and Facebook) are recognized as extremist, their activities are prohibited on the territory of the Russian Federation.
Source link
www.securitylab.ru
