In an effort to establish a possible AI facility, the Netherlands government has gone into talks with AMD and Nvidia, the US chip makers for an agreement to supply hardware and technological knowledge at its proposed AI facility.
According to the country, a potential AI supercomputing facility is part of the Netherlands’ broader strategy to bolster AI research and contribute to EU efforts to strengthen Europe’s digital economy.
The talks bring closer plans by the Netherlands to develop an AI facility
In 2024, the Netherlands government disbursed $210 million for AI investments, with additional funding expected from European subsidies.
According to the Netherlands economy minister Dirk Beljaarts, the Nvidia and AMD talks have increased the project’s chances of becoming a reality. The minister did not give any specific details about the talks.
“Competition is fierce,” said the minister after meeting the two tech giants.
“The whole world is after this technology and this deal brings building a Dutch AI facility a lot closer.”
Beljaarts.
The AI supercomputer facility aligns with the European Union’s broader goals of digital sovereignty and reducing reliance on external technology providers. The Netherlands’ partnership with Nvidia positions the country as a significant player in the AI industry and high-performance computing sector.
The facility is expected to contribute to advancements in AI-driven research, benefiting various industries, including healthcare, logistics, and energy.
The development comes at a time when AMD has just been downgraded by market watchers on fears the tech firm might not stand competition from giants like Nvidia. Morgan Stanley noted the possibility of tough competition from Nvidia to limit the company’s growth opportunities in the AI space was high.
Before markets opened on January 8, HSBC issued a double downgrade for the AMD stock rating as the bank lowered its rating from buy to reduce as well as cut its 12-month price target to $110.
HSBC analysts also expressed concerns over AMD’s slowed-down non-AI data center revenue growth and client momentum, projecting that client revenue would rise by just 12% year on year in fiscal 2025 versus 44% growth in fiscal 2024.
However, the collaboration with the Netherlands government might just be what AMD needs and might offset the projected slowed growth.
The government is keeping a tight lid on project
Although the Dutch government sees vast opportunities in AI, there is still more being kept under wraps about the project. For instance, the government has not yet announced timelines for the project.
This comes as Nvidia CEO Jensen Huang met SK Group chairman Chey Tae-won on Wednesday to discuss AI, suggesting the two companies are exploring ways to deepen one of the most important relationships in AI hardware.
Chey and Huang reportedly discussed areas including moving AI into the physical realm as well as SK Hynix’s high bandwidth memory (HBM) chips, which work with Nvidia’s AI accelerators.
The South Korean company has been quickening its pace of development to make sure it keeps up with Nvidia’s demands for faster evolution, Chey told reporters at the annual CES in Las Vegas.
SK Hynix has become a leading provider of essential memory to Nvidia during the post-ChatGPT AI boom, pushing aside longtime rival Samsung Electronics Co.
Huang has discussed extending AI into more of the physical world to develop tools that can “proceed, reason, plan, and act.” It would transform industries worth $50 trillion, according to Nvidia.
Chey said he told Huang at their meeting that South Korea, a manufacturing powerhouse is uniquely positioned to work with Nvidia on that front.
Both companies have benefited from the insatiable appetite that major companies and governments have for Nvidia’s chips, the gold standard for training AI algorithms. SK Group is said to be focusing on AI data centers to drive future growth, part of a strategy to transform South Korea’s second-biggest conglomerate into a technology-driven business.
SK Hynix’s exhibition at CES highlighted the company’s most advanced HBM chips the 16-layer HBM3E as well as enterprise data storage solutions for data centers.
From Zero to Web3 Pro: Your 90-Day Career Launch Plan