Rome/Milan Jan. 24, 2024 - iGenius and Cineca are creating a new Foundational Large Language Model to help companies and Public Administration leverage the benefits of Generative Artificial Intelligence in key areas such as healthcare, finance, and national security. Guided by the highest compliance, privacy and national security regulations.
iGenius is an Italian company that has been operating since 2016 in the field of research and development of Generative Artificial Intelligence technologies. They have been recognized by Gartner as one of the most dynamic companies globally in the AI Core Technologies area. Cineca is a nonprofit Inter-University Consortium consisting of 118 public entities including 2 ministries and 70 Italian universities, whose mission is to support innovation and promote the use of the most advanced high-performance scientific computing systems. The Leonardo supercomputer, managed by Cineca, is ranked sixth out of the 500 most powerful supercomputers in the world, ranking second in all of Europe.
The agreement signed by iGenius and Cineca will lead to the development of a next-generation Foundational Large Language Model (“Foundational LLM”) - a neural network model that's at the core of Artificial Intelligence tools like ChatGPT or Crystal. This model will be released under an open-source license, trained with several trillion tokens and focusing on businesses and Public Administration, in Italy and Europe.
Reduction of bias and discrimination
A key aspect of the collaboration between iGenius and Cineca is the focus on reducing the biases and discrimination that Artificial Intelligence systems often have. “Modello Italia” will be designed and trained with qualitative and controlled data to ensure that Artificial Intelligence is fair and unbiased. This approach aims to ensure that AI models are free of inherent bias and can be used safely in a variety of applications.
Enhancing security for large-scale use in highly critical environments
Security is a key pillar of this initiative. iGenius and Cineca aim to drive responsible and informed innovation in highly regulated industries. The models will be designed to be robust and reliable to ensure optimal performance in highly complex and sensitive production environments (often called mission-critical). The goal is to provide AI solutions that can be deployed on a large scale with maximum security, contributing to significantly improved decision-making in key areas such as healthcare, finance, and national security.
Regulatory compliance and adaptability
Large Language Models will be designed with a deep understanding of compliance needs, ensuring that AI applications fully comply with current national and international regulations, such as the European AI Act. This is critical to ensure that AI technologies are not only technologically advanced, but also secure, reliable, and legally and regulatory compliant.
Data security and privacy
Special emphasis will be placed on data security and privacy. In an era where data protection is of utmost importance, iGenius and Cineca are committed to developing models that meet the highest standards in terms of cybersecurity and privacy protection, a critical factor in all industries and vital in highly regulated ones.
A promising future for Artificial Intelligence
With this collaboration, iGenius and Cineca are committed to shaping the future of Artificial Intelligence, giving Italy a leading perspective in the field of highly advanced technology.
“Modello Italia” not only represents a significant step forward in the large-scale application of Generative Artificial Intelligence, but it does so with a specific focus on ethics, sustainability and security, making the benefits of Artificial Intelligence accessible to all.
“Language models have the power to democratize knowledge. And this will only happen if every country and every language will be represented by these models. No one better than the Italians can turn their language into superpowers. This is Modello Italia, a Generative AI model representing not just our language but one of the most sophisticated cultures through art, culture and everything we’re so well known for. A prime example of the digital Renaissance, where humans do come first. On behalf of iGenius, I am proud to join forces with Cineca, another example of Italian excellence, to support the country’s ecosystem.” Uljan Sharka, CEO and Founder at iGenius.
"We are proud to be working on this project. We believe Italy can create AI products of excellence. iGenius is ready to work with a new ecosystem of organizations that, thanks to Modello Italia, will give birth to unique solutions for enterprises and Italian, European and global institutions" Stefano Parisse, General Manager at iGenius.
“This is really good news for our country and for Europe. The best way to kickoff 2024. Italy can boast one of the best public supercomputing infrastructures in the world and, thanks to this, today an Italian startup enters the international arena of language models, with a solution that is completely open, reliable and fully compliant with national and European principles. The Italian Model will be an extraordinary enabling factor for businesses and public administration in accelerating digital innovation.” Francesco Ubertini, Cineca President.
“Cineca is a consortium and as such is founded on the power of synergy and collaboration. We are honored to have embarked on this adventure that through the use of the supercomputing infrastructure will be able to develop the first language model for intelligence natively in Italian, so as to allow our country to be an active player in the digital transition.” Alessandra Poggiani, Cineca Director General.
Read more: