The federal judge has allowed a complai to a violation of the publication of Meta on the training of artificial ielligence models. Of course, some of the claims of this case have rejected. In Kadrey VS Meta’s case, writers such as Richard Cadrey, Sara Silverman and Tangsis Kots have claimed that Meta has used their books to teach Llama AI models without authorization and deleted the right to publish information to hide possible violations.
Meta refuses to violate the right to publish
According to the TechcRunch report, Meta claims that the use of these data is within the framework of fair use (FAIR USSE) and has argued that the authors have no legal right to file a lawsuit. US Regional Court Judge Vince Chairia said last moh he probably disagreed with the case, but criticized what he regarded as an exaggerated tone of the Writers’ Legal Team.
He has recely stated that the claim of violating the right to publish is clearly damaged by the coinuation of the lawsuit, and has insisted that the authors have ieionally deleted Meta ieionally deleting the publication manageme information (CMI) to hide copyright laws. He added:
“The sum of these claims creates a reasonable infereial (though not so strong) that Meta deletes the publication of the publication of the publication so that the LLAMA model does not display this information in its output and does not specify that it is trained on the basis of the coe of publishing.”
Judge Beria rejected the California State Comprehensive Law for accessing and fraud by CDAFA; Because the authors were unable to show the meta directly access their computers or servers and only obtained their data (their books).
This case has revealed details of meta dealing with the rules of publishing; For example, the documes provided to the court show that Mark Zuckerberg has personally allowed the LLAMA team to use the work -based works to teach artificial ielligence models, and some meta members have discussed the use of suspicious coe legally for artificial ielligence training.
