Darío Gil, IBM Research: “Este año estará el primer modelo de inteligencia artificial en español con casos de uso”

The joint initiative between the National Supercomputing Center and IBM marks a milestone in the development of artificial intelligence, focused on creating the first large foundational language model in Spanish. This project aims to replicate human reasoning at a large scale, leveraging advanced supercomputing infrastructure to process the linguistic and cultural richness of more than one billion Spanish speakers, thereby establishing a solid, non-transitory technological foundation for the future of the sector. The core of this proposal lies in its collaborative and open architecture, which breaks away from traditional proprietary models. By offering a transparent ecosystem where any university, institution, or community can contribute, it democratizes knowledge creation. This strategy fosters continuous innovation through contextual data adjustment, enabling the system to understand and respect the diverse dialectal variants of Spanish, from administrative usage in Spain to specific expressions in Iberoamerica. This case is fundamental to the open data movement, as it explicitly uses public information, such as parliamentary transcripts and academic resources, as raw material for training. By prioritizing transparency in data and methodology, it demonstrates how public access to government and scientific information can enhance digital sovereignty and equal access to advanced technologies, ensuring that AI development benefits society as a whole rather than being restricted to closed corporate interests.

Source: elpais.com
Published on 2024-04-09

COVERUP: FOIA documents prove 2020 election fraud in Detroit – NaturalNews.com »

Open Data

Related news