Germany’s SPRIND and the Netherlands’ NADI are launching a twenty-month Challenge to utilize AI for the design of high-end AI chips. The goal is to reduce the design phase from years to weeks and create production-ready, energy-efficient chips specifically tailored for AI applications.
The Netherlands and Germany are joining forces to invest together in the development of new AI chips. To this end, the innovation agencies NADI (Netherlands) and SPRIND (Germany) are allocating a sum of 40 million euros.
The organizers aim to radically rethink the entire chip design process by placing AI at the center of every step in the workflow. Instead of classic, lengthy iterations, AI systems must autonomously direct, optimize, and accelerate the process. In doing so, SPRIND and NADI are focusing on both technological progress and European strategic autonomy regarding critical computing infrastructure.
According to the initiators, the Challenge should result in tangible chip designs that can actually evolve toward production. Teams will therefore receive not only financial support but also technical guidance to bring their designs through all design phases up to tape-out preparation.
AI as an autonomous chip designer
Central to the Challenge is the idea that AI is no longer just a tool, but becomes the actual chip designer. From architecture searches to RTL generation and verification, AI systems must autonomously manage the entire development workflow. This includes complex decisions regarding design trade-offs, which currently often occupy large teams for months.
Reinforcement learning and autonomous agent systems play a key role in this. By continuously learning from previous design steps, these systems can seek the best configurations for performance, consumption, and scalability. The organizers expect that design timelines, which traditionally take years, can consequently be compressed into a few weeks. The goal is to set a new standard for high-performance AI chips.
European AI Giga-Factories
The requested chip architectures must not only be fast but also energy-efficient and aligned with specific European needs. SPRIND and NADI explicitly request designs for AI Giga-Factories and domains where Europe has a strong presence, such as data security, biotechnology, robotics, and mobility. This aligns with the desire to reduce technological dependence on China and America.