Europe's AI Computing Power Breakthrough: Germany's Jupiter Supercomputer Unveiled, Triggering a New Landscape of Global Tech Competition

The German Research Center in Jülich officially launched the Jupiter Supercomputer on Friday, representing a technological breakthrough in Europe. According to Reuters, German Chancellor Friedrich Merz personally presided over the inauguration ceremony, witnessing the operation of the most powerful artificial intelligence computing device in Europe to date.

The system is jointly supported by the French IT company Atos and the German company ParTec, currently ranking fourth in the global supercomputer list. Jupiter has become the first exascale supercomputer in Europe to reach the level of 100 billion billion ( operations per second, with impressive performance metrics: it can perform one quadrillion calculations per second, equivalent to the total computing power of 10 million laptops.

Milestone of European Technological Independence

Prime Minister Melitz referred to Jupiter as "a historically groundbreaking project for Europe" at the launch ceremony. He emphasized that this is Europe's technological response to the United States and China, as both countries have taken a leading position in the development of artificial intelligence technology and related industries.

"We in Germany and Europe have all the conditions to catch up and ultimately establish a foothold in the global AI competition," Merz said at the press conference.

The design intention of the Jupiter Supercomputer is to serve scientific fields such as biotechnology and climate research. Its powerful computing capabilities are crucial for running complex simulations, constructing precise models, and processing massive datasets. However, the deployment of this device is also seen as a clear political signal—Europe is actively pushing to reduce its dependence on foreign-controlled digital infrastructure and chips, particularly products from American tech giants and Chinese manufacturers.

The Industry Calls for Open Sharing of High-Performance Computing Resources

Ralf Wintergerst, head of the German Digital Business Association Bitkom, stated that Jupiter will drive Germany to the forefront of the global high-performance computing field and significantly enhance the country's artificial intelligence research and development capabilities.

Wintergest emphasized the importance of resource sharing: "Access to Jupiter should be as simple and convenient as possible for startups and mature companies." This call reflects the common desire of the industry - hoping that this top-notch infrastructure can promote the overall development of the European AI innovation ecosystem.

The European Union is clearly lagging behind in developing the hardware infrastructure needed to support large-scale AI research and development. While Silicon Valley engineers are training large models on racks equipped with dense Nvidia GPUs, and Chinese laboratories are rapidly expanding their computing power through national funding, Europe is essentially on the sidelines. The deployment of Jupiter marks the beginning of a positive change in this situation.

The New Proposal in the United States May Restrict Global Chip Supply

On the same day that the Jupiter Supercomputer went live, Nvidia publicly issued a warning that the United States is brewing a new bill that could hinder future upgrade plans for systems like Jupiter.

The bill is titled the "Guaranteeing Access to and Innovation in Artificial Intelligence Act" ) GAIN AI Act (, and it was proposed as part of the "National Defense Authorization Act." The bill requires artificial intelligence chip manufacturers to prioritize domestic orders over international shipping demands.

A spokesperson for Nvidia expressed concerns, believing that the bill may do more harm than good:

"We will never deprive our American customers of their rights in order to serve other regions of the world. This proposed legislation attempts to address a non-existent problem, and it will limit competition in any industry globally that uses mainstream computing chips."

The bill proposes strict licensing rules: chips with a performance score over 4800 are required to obtain an export license, and the U.S. Department of Commerce will have the authority to completely deny the issuance of licenses. The bill clearly states that if U.S. buyers are still waiting for supply, the export of advanced chips should be halted.

This proposal continues the "Artificial Intelligence Diffusion Rules" implemented by former President Joe Biden, which restrict the performance of processors that U.S. companies can export to other countries. The main goal of these controls is to ensure that the United States has priority in acquiring critical artificial intelligence infrastructure and to prevent China from obtaining high-performance hardware that could be used for military AI development.

Friction Arises in US Technology Regulation Policy

Despite the aforementioned restrictions, U.S. President Donald Trump reached an agreement with Nvidia in August to allow the company to resume exports of certain banned artificial intelligence chips to China. In exchange, the Trump administration will receive a percentage of the sales revenue from Nvidia's export business.

This agreement has raised questions in the industry: do economic interests outweigh safety considerations in the regulation of artificial intelligence technology?

While the United States continues to strengthen its regulations, Germany is striving to open its computing resources—at least within the European Union. Europe's strategic goal is to build autonomous infrastructure capable of supporting the development of the next generation of artificial intelligence, reducing dependence on ongoing permissions from Washington.

However, the challenge is that the vast majority of the world's most advanced AI chips still come from the United States, and Nvidia remains the core supplier for almost all major AI systems globally. This reality makes Europe's path to achieving technological autonomy full of challenges.