El modelo Alia del Gobierno español utiliza obras sin derechos de autor en su entrenamiento de IA, generando controversia legal

The Spanish government’s AI model, Alia, has sparked a debate over the legality of training AI systems on copyrighted works used without authorization. While the government maintains that its practices comply with EU regulations permitting data mining unless rights are explicitly reserved, this stance exposes a significant gap in current open data governance. The existing legal framework, originally crafted for academic research, is ill-equipped to protect creators in the era of generative AI, raising serious concerns about transparency and consent in the datasets used to train these models. Critics argue that effectively reserving rights online is nearly impossible for authors due to the complex technical requirements involved. This structural imbalance means that open access to data for AI development often comes at the expense of intellectual property protections. The situation highlights a tension between fostering innovation through accessible public data and respecting the legal rights of content creators, challenging the assumption that publicly available internet data is free for unrestricted commercial use. This controversy underscores the urgent need for clearer legislative standards in the open data sector. As courts worldwide grapple with similar disputes over unauthorized training data, the resolution of these issues will likely redefine how public and private entities handle copyrighted information. Establishing robust, enforceable guidelines is essential to balance technological progress with the protection of intellectual property, ensuring that open data initiatives do not inadvertently undermine the rights of original authors.

Source: larepublica.es
Published on 2025-09-18

El modelo Alia del Gobierno español utiliza obras sin derechos de autor en su entrenamiento de IA, generando controversia legal

Related news