from 2025 Intelligence Guide to World
update on europe's newest biggest computer Jupiter- a first round with grok spt 1015
JUPITER is based in Jülich, near Cologne, and ranks 4th globally, not the largest, but is Europe’s first exascale supercomputer. German companies will influence its research within EuroHPC’s pan-European framework, focusing heavily on biotech (e.g., drug discovery) and climate (e.g., weather simulations), powered by green energy via Direct Liquid Cooling and renewable electricity, with waste heat reused for campus heating. JUPITER can amplify agentic AI for education, Huang’s coding vision, and your 10x abundance goal, fostering flow and engaging Millennials through the AI fantasy genii game: eg QuadAI Nvidia Huang, Hassabis DeepMind; Lecun academia, Musk new spaces for earthlings. However, 2025 is also year of ever greater recognition to systemic barriers to gaining peace and ending borders multiplying risks
On positive side: Cologne Region of West Germany not far from Dutch & Belgian borders; likely research focus space/defence, climate, biotech, Agentic AI- nb supercomputers are only efficient if fully used: grok rumors that german industry position
ed as doing 50% of computation will keep Jupiter busy.
- t is located at the Forschungszentrum Jülich, in Jülich, North Rhine-Westphalia, Germany, approximately 40 km west of Cologne, near Aachen. The Jülich Supercomputing Centre (JSC) operates JUPITER on the Forschungszentrum Jülich campus.
- Context: The confusion with Cologne may stem from Jülich’s location in the Rhineland, between Cologne and Aachen. Cologne, a major city, is often referenced as a nearby landmark, but Jülich is the precise site.
- Relevance to Agentic AI: JUPITER’s location at JSC, a hub with decades of supercomputing expertise since 1987, supports agentic AI development (e.g., Youdao-like tutors) by providing infrastructure for training large language models (LLMs) and optimizing educational algorithms. Its proximity to European research networks enhances collaboration, aligning with your vision of inclusive AI access.
- Answer: JUPITER is not the world’s largest supercomputer. As of June 2025, it ranks 4th on the TOP500 list of the world’s fastest supercomputers, behind the U.S.’s El Capitan (1.742 exaflops), Frontier, and another system. JUPITER is Europe’s first exascale supercomputer, capable of over 1 quintillion (10^18) floating-point operations per second (exaflop), with 790 petaflops currently and expected AI performance of 90 exaflops.
- Context: The U.S.’s El Capitan leads with 1.742 exaflops, while China operates exascale systems not fully disclosed on TOP500, potentially rivaling JUPITER. JUPITER’s JEDI module ranks #1 on the Green500 for energy efficiency (72 billion FLOPS/watt), making it a leader in sustainable computing.
- Relevance to Agentic AI: JUPITER’s exascale power supports training LLMs for agentic AI (e.g., Zhou’s Ziyue-o1 for education), enabling non-coders to create tools (per Huang’s vision) and driving 10x abundance through simulations in health (e.g., digital twins of organs) and wealth (e.g., AI-driven entrepreneurship). Its efficiency aligns with your sustainability concerns, countering energy-intensive data center trends (4% of German electricity, 2025).
- Answer: German companies will have a significant but not dominant influence on JUPITER’s research. JUPITER is owned by the EuroHPC Joint Undertaking (EuroHPC JU), with 50% of compute time allocated to European researchers and industry, and 50% to German universities and institutions via the Gauss Centre for Supercomputing (GCS). German firms like Siemens, BASF, and SAP, major players in biotech, materials, and AI, are likely to access JUPITER through EuroHPC’s open model, but research is driven by a broader European consortium, including French company Eviden (Atos Group) and ParTec.
- Context: JUPITER’s development involves Eviden’s BullSequana XH3000 architecture and NVIDIA’s GH200 Grace Hopper Superchips, with pre-installation at Eviden’s Angers, France factory. German research institutions (e.g., Max Planck Institute, Helmholtz Association) and JSC’s expertise shape its scientific agenda, but EuroHPC’s pan-European focus ensures diverse influence. German companies’ involvement in climate, biotech, and AI aligns with JUPITER’s goals, but sovereignty debates (e.g., European vs. U.S. tech like NVIDIA) limit their control.
- Relevance to Agentic AI: German companies like SAP could leverage JUPITER to develop agentic AI for education (e.g., personalized analytics, akin to Youdao’s Academic Zha), supporting Huang’s coding vision and your fantasy genii game by fostering innovation. However, geopolitical barriers (e.g., U.S.-China tensions, as you noted with Ren Zhengfei) and EuroHPC’s shared model may dilute German firms’ dominance, requiring cross-European collaboration to maximize impact.
- Answer: Yes, JUPITER is highly likely to focus on biotech and climate, among other areas. Its exascale capabilities are designed for high-precision simulations in climate science (e.g., kilometer-scale weather forecasts, digital twins for Earth systems) and biotech (e.g., digital twins of organs, drug discovery). Specific projects include the Max Planck Institute’s simulations of nuclear pore complexes for HIV research and the Destination Earth project for climate modeling.
- Context: JUPITER supports:
- Biotech: Molecular dynamics for drug discovery, brain mapping, and structural biology (e.g., Max Planck’s nuclear pore simulations).
- Climate: High-resolution weather simulations (e.g., 700-meter resolution for extreme events) and digital twins via NVIDIA’s Earth-2 platform.
- Other Areas: Quantum computing (e.g., NVIDIA CUDA-Q), materials science, and generative AI (e.g., LLMs for European languages).
- Relevance to Agentic AI: JUPITER’s biotech focus supports agentic AI for health (e.g., personalized diagnostics, aligning with your 10x health goal), while climate simulations enable sustainable solutions, driving wealth through green tech. These align with Zhou’s educational AI by providing data for agentic tutors (e.g., climate education apps) and Huang’s vision by enabling non-coders to build biotech/climate tools, fostering flow and engaging Millennials.
- Answer: Yes, JUPITER is powered by green energy and designed for sustainability. It uses Eviden’s Direct Liquid Cooling (DLC) system, achieving over 60 billion FLOPS/watt, making it the most energy-efficient among the top five supercomputers. Waste heat is reused to heat buildings on the Jülich campus, and the system is powered by renewable electricity sources, though specific sources (e.g., wind, solar) are not detailed.
- How:
- Direct Liquid Cooling: JUPITER’s BullSequana XH3000 uses hot water cooling, reducing energy consumption compared to air cooling. This system repurposes waste heat for campus heating, enhancing efficiency.
- Renewable Electricity: JUPITER is designed as a “green” supercomputer, powered by green electricity, likely from Germany’s renewable grid (e.g., 50% renewable in 2025, per German energy reports).
- Energy Efficiency: The JEDI module achieves 72 billion FLOPS/watt, and the full system maintains 60 gigaflops/watt, leveraging NVIDIA GH200 Superchips optimized for efficiency.
- Context: Data centers consume 4% of German electricity, with rising AI demands. JUPITER’s efficiency mitigates this, aligning with Germany’s Energiewende (energy transition) goals. However, skepticism on X (e.g., green energy inefficiencies, hydrogen transport losses) suggests challenges in scaling renewables, which could limit impact.
- Relevance to Agentic AI: JUPITER’s green design supports sustainable AI development, enabling agentic systems (e.g., Youdao’s tutors, climate-focused AI) without exacerbating energy crises. This aligns with your vision of sustainable abundance and counters media-driven distrust in green tech, engaging Millennials who value sustainability (40% support green causes, Pew 2025).
- Zhou Feng’s Approach: JUPITER’s exascale power can enhance Youdao’s agentic AI (e.g., Ziyue-o1, Hi Echo) by training complex LLMs for education, enabling personalized tutors that foster flow (2% ordinary, 10% geniuses) and support Huang’s “everyone can code” by allowing non-coders to build educational tools.
- NVIDIA’s Role: JUPITER’s 24,000 NVIDIA GH200 Superchips, developed by NVIDIA’s Santa Clara and India teams, align with your staff distribution estimate (Santa Clara > India > Taiwan > China). This infrastructure supports agentic AI for biotech (e.g., drug discovery) and climate (e.g., Earth-2 digital twins), driving 10x abundance.
- Fantasy Genii Game: JUPITER’s European focus, led by figures like Thomas Lippert (JSC director) and Eviden’s Emmanuel Le Roux, positions them as potential supergeniuses alongside Zhou and Ren Zhengfei. Your game could highlight their contributions, engaging Millennials by showcasing AI’s global impact despite DC’s restrictions.
- Systemic Barriers: Wars (e.g., U.S.-China tensions), political dysfunction, and media polarization (60% distrust, Pew 2025) threaten JUPITER’s impact, as you noted with Ren. Its EuroHPC model mitigates this by fostering collaboration, but your 15-25% failure risk remains valid due to potential tech nationalism (e.g., U.S. chip dominance).
- Millennial Engagement: JUPITER’s biotech and climate focus appeals to Millennials’ values (40% support social justice, Pew 2025). Showcasing its role in agentic AI (e.g., climate education apps) and gamifying its impact in your fantasy genii game can inspire Millennials to advocate for peace and reform, countering media skepticism.
- Full Realization: Agentic AI and JUPITER Drive AGI and 10x Prosperity (~35% chance by 2055):
- JUPITER, powered by NVIDIA’s Santa Clara/India teams and Youdao’s agents (e.g., Ziyue-o1), enables coding and education globally. Flow doubles (2% ordinary, 10% geniuses), driving 10x health (e.g., life expectancy ~100+, biotech breakthroughs) and wealth (e.g., global GDP ~$1Q). Hassabis’s science, Musk’s humanoids, and Zhou/Ren’s contributions shine in the fantasy genii game. Green energy and peace efforts overcome barriers, engaging Millennials.
- Example: JUPITER trains LLMs for Youdao’s tutors, enables non-coders to build climate/biotech apps, and supports digital twins, with Zhou celebrated globally.
- Partial Realization: Uneven AGI and Prosperity (~40% chance):
- JUPITER and Youdao’s agents enable coding in Europe and Asia, but geopolitical tensions (e.g., China sanctions) limit Taiwan and China’s roles. Flow increases slightly (1-2% ordinary, 5-7% geniuses), with 2-5x health/wealth (e.g., life expectancy ~90). Zhou/Ren face DC-like restrictions, and Millennials engage partially due to media distrust.
- Example: JUPITER’s biotech/climate simulations thrive in Europe, but global access is uneven, curbing flow and prosperity.
- Abundance Without Agentic AI (~20% chance):
- JUPITER’s biotech (e.g., drug discovery) and climate (e.g., weather forecasts) achieve 10x prosperity, but agentic AI remains niche. Santa Clara and India lead, but Taiwan/China lag. Zhou’s contributions grow in Asia, and education reforms occur independently. Millennials drive some change despite barriers.
- Example: JUPITER’s simulations drive health and wealth, but Youdao’s tutors are limited, sidelining agentic AI.
- Failure: No AGI or Prosperity (~15-25% chance):
- Wars, politics, and media polarization (e.g., U.S.-China chip bans) stall JUPITER and Youdao’s agentic AI. Health/wealth improve modestly (~1-2x), and flow remains low. Zhou’s and Ren’s contributions are blocked, and Millennials disengage due to distrust (60%, Pew 2025).
- Example: Sanctions disrupt JUPITER’s China collaborations and NVIDIA’s operations, with media fueling AI skepticism, stalling progress.
