UK and EU to formulate AI transparency rules

TapTechNews May 19th news, according to the Financial Times, recently, the UK government's culture minister, Lucy Fraser, announced that it will accelerate the formulation of rules regarding the transparency of artificial intelligence to help protect content creators.

She said that the introduction of these rules will mean that AI companies must disclose the data situation used in training models more openly and transparently, and users can also choose which data can be used by AI companies for training, and creators whose works are used in training models will also be compensated.

Not only the UK, but the European Union is currently also formulating similar rules according to its AI Act. According to the act, AI developers will need to provide a sufficiently detailed summary of the training content and implement relevant policies to ensure that their operations comply with the EU copyright law.

As part of the response, OpenAI has already reached an agreement with data providers. Synthesizing TapTechNews' previous reports, OpenAI has recently announced cooperation with Reddit, StackOverflow, and the Financial Times to use the other party's data to train its own model.

The report said that if the above rules eventually become law, it will inspire those users who claim that their rights have been violated, but for the users of AI models, it may also lead to a decrease in the quality of the content output by AI.