Journal of
Italian
Banking
Association

year 100
 

Archive » May 2025 » Ai and copyright: governing the training of Llms to protect corporate assets

Ai and copyright: governing the training of Llms to protect corporate assets

Tommaso Grotto
May 2025 - n. 5
Keywords: Copyright, Intelligenza artificiale
Jel codes: K11, K24, O30, O34

Generative Ai and multimodal Large Language Models are posing important questions on the copyright front, forcing professionals, companies and legislators around the world to reflect on the evolution of such an important shake-up. Of great concern is not only the ability of generative AI to manipulate content, but also the issues in relation to the authorship of the works created and the exploitation of pre-existing works for model training. The question is also economic, since the data fed to the Llms is the result of huge investments. If these data then come from pirated content, the problem becomes even more complex. Generative AI is introduced into a scenario already compromised by unfair competition, counterfeiting, fraud, and will necessarily require a clear stance on the part of copyright holders.

Interested in this paper?
Buy the issue