The first generation of FDX Fusion is expected to achieve digital performance on par with 7-nanometer-class technology. GF is developing the technology from the ground up—as the next major step in its FDX™ FD-SOI roadmap. To achieve this, the platform combines advanced substrate engineering, front-end transistor innovations, and back-end interconnect and integration technologies with high-performance RF, analog, and mixed-signal capabilities. It also features exceptionally low leakage current, in-memory computing, embedded non-volatile memory, and high logic density. Compared to the first generation of FDX, FDX Fusion is expected to achieve more than double the density. This provides customers with a comprehensive, intelligent platform for edge and physical AI applications that simultaneously reduces system complexity, latency, and power consumption.
“Physical AI is creating a new class of requirements for semiconductors. Intelligence is moving from the cloud into systems and devices that process data in real time and interact seamlessly with their physical environment,” says Ed Kaste, Senior Vice President of GF’s CMOS business. “FDX Fusion expands our FD-SOI roadmap with innovations across the entire technology stack—from the substrate to the transistor to the interconnect. This creates a platform specifically designed for energy-efficient Physical AI applications. Combined with our software-to-silicon expertise and our growing, resilient manufacturing base in Europe, we aim to position GF with FDX Fusion as the technology partner of choice when it comes to scaling physical AI applications faster.”
The new platform builds on GF’s long-standing leadership in FD-SOI technology and energy-efficient semiconductor design for secure, connected devices. Billions of these chips are already in use today in a wide range of applications, including radar systems and sensors for vehicles, wireless microcontrollers, display drivers, and front-end modules for SATCOM and 5G devices.
FDX Fusion is designed to cover a broad spectrum of physical AI systems and applications in the automotive, consumer, and industrial sectors—ranging from edge inference to robotics and industrial automation, as well as autonomous systems, sensor fusion architectures, and smart connected devices. GF is already working with several customers on the early development and evaluation of FDX Fusion. The requirements of real-world applications are directly incorporated into the product design. This is intended to shorten the path from platform definition to series production.
To help customers bring their FDX Fusion designs to market faster, GF’s comprehensive design enablement ecosystem is available to them. This includes the advanced MIPS and ARC portfolio with workload-oriented processor IP, the RISC-V instruction set architecture (ISA), and expertise in application-specific processor design optimized for Physical AI. These resources are intended to complement the FDX Fusion platform with scalable, standards-based solutions across the entire Physical AI signal chain. This enables GF to support its customers at every stage of the design cycle—from the development of the computing architecture to high-volume mass production.
The first production run of FDX Fusion is planned at GF’s Dresden facility. There, GF—supported under the European Chips Act—is currently expanding its capacity for differentiated semiconductor solutions. These are of central importance for security and defense, industrial competitiveness, resilient supply chains, and Europe’s technological sovereignty. Through this expansion and its strategic technology roadmap, GF aims to strengthen Europe’s leading role in the development and manufacturing of key technologies—for AI-powered industrial systems, secure connectivity, and next-generation intelligent systems.
The first demonstrator chips based on FDX Fusion are expected to be available to customers for evaluation as early as the beginning of 2027. An initial Process Design Kit (PDK) is planned for mid-2027. In addition, the platform is expected to offer optimized process flows and cost structures that enable rapid prototyping and a shorter time-to-market.
“We are very pleased that GlobalFoundries has announced its next-generation FD-SOI node in the single-digit nanometer range. It builds on a technology that CEA-Leti has been advancing for decades. With the FAMES pilot line, CEA-Leti is actively shaping the future European FD-SOI roadmap and works every day to establish the technology in as many applications as possible,” explains Sébastien Dauvé, CEO of CEA-Leti.
“Europe’s economy runs on silicon—wafer by wafer. Chips are crucial for applications such as robotics and connectivity, and for bringing AI into the physical world and into industry. Realizing this vision requires close collaboration between the semiconductor industry, industrial automation, and software ecosystems. GF’s FDX Fusion Roadmap strengthens the technological foundation for Physical AI. It advances capabilities that will enable smarter, more efficient, and more resilient industrial systems,” says Cedrik Neike, Member of the Executive Board of Siemens AG and President of Digital Industries.
“The development and manufacturing of high-performance computing modules for edge processing are key to the next generation of advanced robotics and autonomous systems. GF’s technology roadmap represents exactly the kind of semiconductor investment needed to bring Physical AI to scale,” says Karn Budhiraj, Vice President of Supply Chain at Tesla.
About GF
GlobalFoundries (GF) is one of the world’s leading manufacturers of semiconductors that are essential for scalable AI solutions, from the cloud to physical applications. Through close, trusting partnerships with its customers, the company delivers differentiated, energy-efficient, and high-performance solutions for the automotive, aerospace, defense, data center, smart mobile device, Internet of Things, and other high-growth markets. With manufacturing facilities in the U.S., Europe, and Asia, the company is a globally recognized, end-to-end technology partner. GF’s talented, international team works every day with a focus on safety, durability, and sustainability.
Forward-Looking Statements
This press release may contain forward-looking statements that involve risks and uncertainties. Readers should therefore not place undue reliance on these statements. They are valid only as of the date of publication. GF is under no obligation to update these forward-looking statements to reflect events or circumstances after the date of this press release or actual results, unless required by law.
– – – – –
Related Links
👉 www.gf.com
Photo: GlobalFoundries
