The European Union (EU) has revealed its plans to make changes to the existing regulation that governs the European High Performing Computer Joint Undertaking (EuroHPC), a body responsible for overseeing the use of supercomputers in artificial intelligence (AI) development.
On May 24, the EU updated the regulation to include the objective of establishing and operating local AI factories. The EU believes that such an initiative would provide additional infrastructure for AI supercomputing services.
The proposed amendment, which was put forward by the commission, builds upon the foundational work of EuroHPC, an organization established in 2018 to spearhead European supercomputing initiatives.
Currently, EuroHPC manages nine supercomputers located throughout Europe, which have been developed since the inception of the initiative. The introduction of AI factories, as suggested by the amendment, will further contribute to this mission.
Furthermore, an increased capacity for supercomputing within the EU would provide AI startups with better resources to innovate and train their models.
Willy Borsus, the vice-president of Wallonia and the minister for economy, research, and innovation, emphasized the collective effort required to achieve this significant milestone.
The council has called for the EuroHPC Governing Board to establish special access conditions for these AI supercomputers, with a priority given to startups and small to medium-sized enterprises.
According to the new regulation, hosting entities can receive up to 50% of the acquisition and operational costs of AI supercomputers from the EU. Ownership of these machines can be transferred to the hosting entities after five years of acceptance testing.
The primary focus of the AI supercomputers will be on the development, testing, evaluation, and validation of large-scale AI training models and emerging AI applications, thereby advancing AI solutions within the EU.
Before taking effect, the proposed regulation must undergo legal review, and upon approval, it will be published in the Official Journal of the European Union, with a 20-day period before it becomes effective.
The commission’s proposal is part of a broader AI initiative introduced by the EU’s president, Ursula von der Leyen. It represents a strategic move to enhance Europe’s AI capabilities.
The EU has been at the forefront of developing AI policies for implementation across the region. In March 2024, the EU passed the groundbreaking AI Act, the world’s first comprehensive set of regulations aimed at ensuring safe and ethical AI development in the region.
In a related matter, sci-fi author David Brin suggests using AI to prevent an AI apocalypse by having AI systems monitor and regulate each other.