The letter to ministers calls on the government to improve transparency around training data to enable creators to identify the use of their works, ensure that any copyright holders who opt out aren’t penalised through reduced visibility, and enable consumers to make informed choices about which AI model to use.
The MPs argued there needed to be more transparency around the vast amounts of data used to train generative AI models, and urged the government not to press ahead with plans to require creators to opt out of having their data used.
The government’s preferred solution to the tension between AI and copyright law is to allow AI companies to train the models on copyrighted work by giving them an exception for “text and data mining”, while giving creatives the opportunity to opt out through a “rights reservation” system.
She added: “Aside from any changes to copyright, there needs to be much tougher requirements on transparency of the data being used to train AI models, so creators will know without ambiguity where they need to be remunerated for the use of their works.”.
The culture, media and sport committee and science, innovation and technology committee were responding to a government consultation on artificial intelligence and copyright, after their joint evidence session held earlier this month with representatives from AI startups and the creative industries.
