As big tech lobby weakens the AI bill in Brazil, RSF calls on the Senate to reverse changes to text
The approved report on Brazil’s AI bill completely excluded artificial intelligence (AI) systems used to moderate and recommend content on social networks from the scope of the law. Reporters Without Borders (RSF) calls on all senators to reverse this setback in the vote on 10 December.
On 5 December, the Temporary Committee on Artificial Intelligence approved Senator Eduardo Gomes' report on the bill regulating the development and use of AI systems in Brazil. The text makes clear progress by including the integrity of information among the foundations of the law, and by protecting and promoting the reliability, accuracy and consistency of information in order to strengthen fundamental rights. It also establishes remuneration for authors, creators and journalists for the use of copyrighted content in the training of AI systems - as when ChatGPT bases its answers on journalistic content.
However, due to strong pressure from digital platforms and far-right parties, AI’s role in moderating and recommending content has been excluded from the scope of the law, meaning these systems will remain completely unregulated.
“The Brazilian law could pioneer the protection of information integrity and the safeguarding of journalists' copyrights. However, it would be a huge step backwards to not cover the AI systems that play the most structural role in the dissemination of information in Brazil. On 10 December, senators will have the opportunity to reverse this situation in a plenary vote, and regulate social media in order to protect the public's right to reliable information. It is unacceptable that the world's richest companies — who have caused so much damage to public debate in the information space — can trample over democracies to further their own interests in legal processes like AI regulation.
Technology companies have also teamed up with Brazilian industry to try to remove the chapter on respect for copyright from the text. The chapter establishes that creators have the right to be informed if their content has been used in the development of AI systems; to prohibit the use of content they own; and to be duly remunerated for this use. So far, the bill's rapporteur — with the support of the government and civil society and journalism organisations, including RSF — has kept the chapter in the bill. But it runs the risk of being excluded in the plenary vote.
