Various known media organizations have signed a petition to set new rules that will protect copyright in data as used in training generative AI models.
Read Also : AFP Files Legal Action Against X
Conflicts Between AI and Media
In an open letter, several media outlets explained that while the community is fairly accepting of AI and generative models, these technologies have the power to endanger the media ecosystem if left unchecked. The letter cited that the impact of AI may lead to the decline of public's trust in the "independence and quality of content." Moreover, this will also greatly affect the financial stability of creators.
Generative AI and large language models have the ability of producing synthetic content based on the data gathered around its ecosystem. The collection of data where the source can either be factual or fake, can lead to distortion of facts, leaving the public confused on what is true and what is not.
The news outlets also cited that this act is an indeed violation of copyright law hence, a regulation should be put in place to protect the intellectual property of media producers.
Media Demands for AI Regulation
The following are the suggested regulations: include transparency on all the training materials used to create AI models, consent of the intellectual property rights, and give media companies a chance to negotiate about data usage. In addition, AI models should be required to eliminate bias and misinformation, and have AI models be labeled properly as AI-generated content.
As for labeling, some tech companies have already committed and shown that they are working on making it possible for their platforms, as for the others it remains uncertain on how AI companies will react to these demands.
Among the signatories include Agence France-Presse, the European Pressphoto Agency, the European Publishers' Council, Gannett, Getty Images, the National Press Photographers Association, the National Writers Union, News Media Alliance, The Associated Press, and The Authors Guild.
Read Also: TikTok Adds New Feature That Labels AI-Generated Content