Markets US

Samsung, alongside Somerset Capital Partners and other investors, led a $230 million funding round for Dutch AI chipmaker Euclyd, signaling a growing shift toward alternatives to Nvidia’s dominance in AI inference chips. The investment underscores intensifying competition in the GPU market, as companies seek energy-efficient solutions to challenge Nvidia’s near-monopoly in high-end AI hardware.

Dutch startup Euclyd, founded in 2024, raised $230 million in a Series A round co-led by Samsung, Somerset Capital Partners, the Scaleup Europe Fund, and Innovation Industries. The funding includes a €200 million investment aimed at developing AI inference chips with a distinct architecture from Nvidia’s GPUs. Euclyd’s technology targets energy-efficient, self-hosted AI solutions, positioning it as a direct competitor to Nvidia’s dominant market position.

The investment reflects broader industry trends as tech giants and startups race to develop proprietary AI hardware. Samsung’s involvement highlights its strategic push to shape AI infrastructure, leveraging its expertise in memory manufacturing and supply chain logistics. Euclyd’s CEO, Bernardo Kastrup, emphasized that the funding will enable the company to scale its silicon systems, which claim to reduce data center energy costs by up to 70% compared to traditional GPU-based solutions.

Euclyd’s business model combines hardware sales to enterprise clients seeking secure self-hosted AI inference and IP licensing for companies building custom chips. The startup aims to deploy physical chip systems by 2028, with a target of serving thousands of enterprise customers by 2030. However, its commercial scalability remains unproven, as its systems have yet to demonstrate large-scale deployment. Competitors like Google, AWS, and Meta are also advancing their own AI chip projects, including OpenAI’s Jalapeño, which claims industry-leading speed and efficiency.

Samsung’s stake in Euclyd could bolster its supply chain and engineering capabilities, while Euclyd’s focus on energy efficiency may attract cost-conscious enterprises. Risks include potential delays in scaling and competition from established players. Opportunities lie in capturing market share through secure, low-cost AI infrastructure, though Euclyd’s ability to monetize its IP effectively remains a critical uncertainty.

Euclyd plans to begin rolling out physical chip systems in 2028, with a goal of serving thousands of enterprise customers by 2030.

Samsung’s $230 million investment in Euclyd highlights a growing challenge to Nvidia’s AI chip dominance, but the startup’s commercial scalability and ability to compete with tech giants like Google and Meta remain key uncertainties. The move signals a broader industry shift toward diversified AI infrastructure, with energy efficiency and enterprise adoption as pivotal factors.

Topics: AI Technology, Semiconductor Industry, Venture Capital, AI Infrastructure, Data Center Innovation, Enterprise Technology, Chip Manufacturing, AI Hardware, Financial Investment, Tech Innovation

#AIChipRace #SemiconductorInnovation #SamsungInvestment #GPUAlternatives #DataCenterEfficiency #NvidiaCompetitor #EnterpriseAI #TechInnovation #AIInfrastructure #AIHardware

Source: CNBC