Scaleway will build an AI supercomputer in Sweden, its first outside France, running on AI chips from AMD, Reuters reported on 13 October 2026, citing the French cloud company’s announcement at its annual AI conference in Paris.
Scaleway said the machine would be the first AMD-based AI supercomputer for commercial use in Europe, open to startups, AI companies and businesses across the continent, and that it will run on AMD Instinct MI355X chips, according to Reuters.
Business Pill · SWITCHING COSTS
A one-minute explainer of switching costs: what it costs a buyer to move from one supplier to another. It teaches the general idea only and says nothing about any organisation in this story.
The key insight: As we read it, the software point comes from AMD: Stephanie Dismore, AMD’s EMEA head, tied the system to “our open ROCm software ecosystem,” as Reuters reported, and Reuters notes ROCm runs the AI tools that also work on Nvidia chips while CUDA is proprietary.
According to Reuters, the system will run on AMD Instinct MI355X, which the news agency describes as the chipmaker’s answer to Nvidia’s Blackwell line. AMD’s product page lists the MI355X with 288 GB of HBM3E memory and 8 TB/s of peak memory bandwidth.
“Scaleway is expanding the choice of AI infrastructure available to customers across Europe, through AMD Instinct GPUs and our open ROCm software ecosystem,” Stephanie Dismore, AMD’s senior vice president for EMEA, said, as Reuters reported.
What Reuters’ report of 13 October 2026 states about Scaleway’s Swedish AI supercomputer (orange) and what it does not give (grey).
Why the Software Matters
Reuters explains the difference this way: AMD’s ROCm is open-source and runs the widely used AI programming tools that also work on Nvidia chips, while Nvidia’s equivalent, CUDA, is proprietary and runs only on Nvidia chips, locking developed applications to the hardware.
Reuters also notes that AMD chips already power supercomputers used for European research, including LUMI in Finland, and that US cloud firms including Oracle and Vultr rent AMD chips to European customers.
Why Sweden
Scaleway said it picked Sweden because of its largely low-carbon electricity and its AI sector, Reuters reported. Scaleway entered the Swedish market in May through a deal with telecom operator Tele2 to sell its cloud and AI services there.
Tele2’s own release of 28 May 2026 describes that partnership as a sovereign and scalable cloud and AI offering in Sweden, combining Scaleway’s public cloud with Tele2’s existing private cloud services. Tele2’s business-to-business head Stefan Trampus said the supercomputer “sends a clear signal” on Sweden’s attractiveness for AI infrastructure, according to Reuters.
Companies named in Reuters’ report of 13 October 2026, with Tele2 from its 28 May 2026 release; AMD, the chip supplier, is highlighted.
Who Is Building It
Reuters reported that the machine will be assembled and maintained by French firm Bull, with servers from Giga Computing, networking from HPE, storage from VAST Data and cooling liquid from Motul.
The Structural Reading
As we read it, the choice is about the software layer as much as the chip: AMD presents the system through its open ROCm software ecosystem, in Dismore’s words, and Reuters frames CUDA as the proprietary layer that ties applications to Nvidia hardware.
As we read it, a European cloud building its first supercomputer outside France on AMD rather than Nvidia puts a second hardware ecosystem in front of its customers; the texts we read do not measure how many of them use it.
The Structural Read
Reuters names the suppliers: French firm Bull assembles and maintains the machine, with servers from Giga Computing, networking from HPE, storage from VAST Data and cooling liquid from Motul.
Tele2’s May release frames the Swedish partnership around sovereignty: it describes a sovereign and scalable cloud and AI offering in Sweden that combines Scaleway’s public cloud with Tele2’s private cloud services.
As we read it, for AMD the deal adds a commercial European cloud to research systems such as LUMI in Finland, which Reuters says already runs on AMD chips.
Stephanie Dismore, AMD senior vice president for EMEA, as reported by Reuters, 13 October 2026
“Scaleway is expanding the choice of AI infrastructure available to customers across Europe, through AMD Instinct GPUs and our open ROCm software ecosystem”
Three Implications
EUROPEAN AI STARTUPS AND COMPANIES Scaleway says the machine will be open to startups, AI companies and businesses across the continent, according to Reuters.
NVIDIA Reuters describes the MI355X as AMD’s answer to Nvidia’s Blackwell line and CUDA as proprietary software that ties applications to Nvidia hardware.
SWEDEN Scaleway said it picked the country for its largely low-carbon electricity and its AI sector, Reuters reported.
The framework puts it this way: “Close the SCARCE layer (keep proprietary) + Open the ABUNDANT layer (commoditize).”
As we read it, AMD’s pitch for the system leans on the open layer: Dismore names its open ROCm software, which Reuters says runs the widely used AI tools that also work on Nvidia chips.
What Is Not Established
The Reuters report we read gives no GPU count, power capacity, investment figure, start date, exact site or named customers. That the machine is the first AMD-based AI supercomputer for commercial use in Europe is Scaleway’s claim, as Reuters reported it. We did not find Scaleway’s own release on its website or iliad’s on 13 October 2026, so the details here come from Reuters, Tele2’s May release and AMD’s product page.
Scaleway will build its first AI supercomputer outside France in Sweden on AMD Instinct MI355X chips, Reuters reported on 13 October 2026, with Bull assembling it and partners including HPE and VAST Data. Size, cost and timing were not given.
A note on sourcing. This piece is based on Reuters’ report of 13 October 2026 by Leo Marchandon, read in full through a full-text syndication, Tele2’s press release of 28 May 2026 and AMD’s Instinct MI355X product page, read on 13 October 2026. We did not find Scaleway’s own release on its website or iliad’s that day. Nothing here is a forecast, and nothing here is financial or investment advice.
Gennaro Cuofano is a CRO and tech executive who has worked in AI since late 2015, starting with bringing NLP, natural-language generation, voice and chatbot products to the marketing industry. His background is in law and finance: he holds a Master's degree in Law and an International MBA with an emphasis on corporate finance (LUISS Business School and the University of San Diego, 2012), and worked as a financial analyst at a real-estate investment firm in San Diego and as an assistant controller. His work focuses on business model strategy, business engineering and, more broadly, structural analysis: how companies make money, read from their own filings. He created FourWeekMBA and leads research there and at The Business Engineer, his newsletter on AI and business strategy, with over 95,000 subscribers and more than 1,000 published analyses. His writing has also appeared on Entrepreneur, HackerNoon and Search Engine People. Find him on LinkedIn and Substack. See how we source.
Scroll to Top
Discover more from FourWeekMBA
Subscribe now to keep reading and get access to the full archive.
