Artificial Intelligence and Copyright: The EU Should Preserve the Copyright Directive’s Delicate Balance to Safeguard and Promote Innovation

In the past years, new technological developments in AI have attracted widespread political attention and regulatory scrutiny worldwide. In particular, the emergence and rapid adoption of generative AI systems able to generate content such as text, audio, video, and computer code based on users’ instructions triggered heated legal and political discussions, including in the field of copyright. These systems are usually powered by so-called foundation models, which are ‘models trained on broad data (generally using self-supervision at scale) that can be adapted to a wide range of downstream tasks’ . Examples of such models are OpenAI’sGPT-4, Anthropic’s Claude 2, Google’s Gemini, Meta’s Llama 2, Amazon’s Titan or Mistral AI’s Mistral 7B model.

Lees verder:

• Artificial Intelligence and Copyright: The EU Should Preserve the Copyright Directive’s Delicate Balance to Safeguard and Promote Innovation door Boniface de Champris in Journal of AI Law and Regulation