![](\"https:\/\/cdn.pblzr.de\/dacbb27c-1270-4041-b681-e2b95f06f8a1\/2026\/09\/sprind-logo-400x300_TEXT.jpg\"){kind=logo}
<\/p>\n<\/p>\n
Artificial intelligence is rapidly permeating every sector, from medicine and mobility to energy and data security. What is considered an innovation today will be standard tomorrow. The pace of development is accelerating week by week.<\/p>\n
But while the AI landscape is transforming daily, the hardware cannot keep up. The chips intended to power AI systems are designed for yesterday\u2019s models. The scaling rules that have defined semiconductor development for decades no longer hold. Instead, the path to innovation is shifting toward new approaches and bold architectures that go beyond legacy systems.<\/p>\n
The opportunity: New AI models and open-source technologies are lowering the barriers to entry and opening up design possibilities that were unthinkable just a short time ago. The search space for forward-looking ideas is growing. That is why now is the time for a strategic fresh start in chip design\u2014Made in Europe.<\/p>\n
A collaborative project between the European innovation agencies SPRIND (based in Leipzig, Saxony) and NADI (National Agency for Disruptive Ideas in the Netherlands), which is currently being established.<\/p>\n
From architecture search through RTL generation to verification, AI systems are intended to control and optimize the development process largely autonomously. Reinforcement learning and autonomous agent systems make it possible to make complex design decisions independently and reduce development time from years to just a few weeks. AI is becoming the actual chip designer.<\/p>\n
The challenge seeks high-performance and energy-efficient chip architectures specifically tailored to the requirements of AI giga-factories and European areas of strength such as data security, biotechnology, robotics, and mobility. The goal is not incremental improvements to existing systems, but rather to set new benchmarks in computing power, energy efficiency, and scalability.<\/p>\n
The challenge will run for 20 months and consist of two stages. A jury of recognized experts will support SPRIND and NADI in selecting and evaluating the participating teams. SPRIND and NADI offer financial support in the form of individual fixed-price contracts based on the teams\u2019 cost estimates, with up to \u20ac9,600,000 per team across two phases. In Phase 1 (8 months), up to \u20ac2.6 million per team, and in Phase 2 (12 months), up to \u20ac7.0 million per team.<\/p>\n
To help the teams reach their full potential, SPRIND and NADI provide additional support opportunities throughout the entire duration of the challenge. They help facilitate access to potential new collaboration partners and experts. After the end of the second phase, the goal is to secure follow-on funding in the tens of millions for outstanding teams to enable the implementation of the developed designs during the tape-out phase. Specific details regarding potential follow-on funding will be published at a later date.<\/p>\n
The application deadline is November 30, 2026, at 11:00 a.m. (CET). All applications received by this time will be considered.<\/p>\n
– – – – –<\/p>\n
\ud83d\udc49 www.sprind.org\/<\/a> Photo: unsplash<\/i><\/p>\n","protected":false},"excerpt":{"rendered":" September 24, 2026. The goal of the challenge is to use AI to drastically shorten the design process for high-performance chips and to develop concrete, production-ready chip designs for AI applications.<\/p>\n","protected":false},"author":3,"featured_media":0,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"inline_featured_image":false,"footnotes":"","_links_to":"","_links_to_target":""},"categories":[4818],"tags":[1961,1954,4783],"class_list":["post-546187","post","type-post","status-publish","format-standard","hentry","category-microelectronics","tag-ic-design","tag-research-development","tag-semiconductors"],"acf":[],"yoast_head":"\n
\ud83d\udc49 https:\/\/www.sprind.org\/taten\/challenges\/ai-native-chip-design<\/a> &<\/p>\n