Samsung Electronics and Broadcom signed a memorandum of understanding (MOU) to expand their strategic collaboration across memory, foundry, and advanced packaging technologies for next-generation AI infrastructure. The agreement, announced at the AI Summit in San Francisco, outlines collaboration estimated at more than $200 billion over the next five years through 2030, spanning memory supply and advanced semiconductor manufacturing.
The expanded relationship covers several key technology areas. Samsung plans to supply Broadcom with High Bandwidth Memory (HBM) for future AI accelerators, while the companies will also broaden their foundry engagement around Samsung’s 2nm and future process technologies. Broadcom intends to utilize Samsung’s advanced manufacturing capabilities for products including Wireless Broadband Communications (WBC) solutions. The collaboration also includes advanced packaging technologies such as 2.3D and 2.5D integration built around Samsung’s 2nm manufacturing platform, supporting higher-performance and more power-efficient AI and networking processors.
The announcement underscores the increasing integration between memory suppliers, foundries, packaging providers, and AI silicon designers as demand for accelerated computing continues to grow. Samsung said its combination of memory, logic, foundry, and advanced packaging capabilities positions it to provide end-to-end semiconductor technologies for AI infrastructure, while Broadcom continues expanding its portfolio of AI accelerators, networking silicon, custom ASICs, and connectivity solutions for hyperscale cloud deployments.
| Samsung–Broadcom AI Semiconductor Collaboration | |
| Agreement | Memorandum of Understanding (MOU) announced during the AI Summit in San Francisco. |
| Estimated Value | More than $200 billion in collaboration across memory and foundry technologies through 2030. |
| Memory | Strategic collaboration on High Bandwidth Memory (HBM) supporting Broadcom’s next-generation AI accelerators. |
| Foundry | Expanded manufacturing collaboration using Samsung’s 2nm process technology for Broadcom semiconductor products. |
| Product Focus | Includes Wireless Broadband Communications (WBC) devices. |
| Advanced Packaging | Joint work on 2.3D and 2.5D heterogeneous integration technologies. |
| End Markets | Designed for AI infrastructure, high-performance computing, and next-generation networking silicon. |
| Technology Scope | End-to-end collaboration spanning memory, logic manufacturing, and advanced packaging for future AI platforms. |
“As AI infrastructure continues to scale, close collaboration across the semiconductor ecosystem becomes increasingly important,” said Charlie Kawwas, President, Semiconductor Solutions Group at Broadcom. “By combining Samsung’s memory and foundry expertise with Broadcom’s AI and connectivity leadership, we aim to continue to deliver technologies that power the next generation of AI infrastructure.”
🌐 Analysis
The agreement reflects a broader industry trend toward vertically integrated semiconductor supply chains for AI infrastructure. AI platforms increasingly depend on close coordination between advanced logic manufacturing, HBM, advanced packaging, and high-speed networking to deliver competitive performance and power efficiency. Foundries capable of combining these technologies are becoming increasingly important partners for AI chip developers.
Broadcom continues expanding its AI infrastructure business through custom accelerators, Ethernet networking silicon, PCIe connectivity, optical interconnects, and switching platforms. Samsung, meanwhile, is investing heavily in advanced foundry nodes, HBM, and heterogeneous packaging as it competes for leading-edge manufacturing alongside other global semiconductor foundries.
Related Converge Digest Coverage Broadcom’s major AI silicon, hyperscaler and semiconductor supply agreements | |
JULY 8, 2026 Semiconductors | Apple Commits $30 Billion-Plus to Broadcom for U.S.-Made RF Chips Apple expanded its long-term U.S. semiconductor sourcing relationship with Broadcom, covering RF components manufactured through Broadcom’s domestic supply chain. |
JUNE 24, 2026 Semiconductors | OpenAI and Broadcom Unveil “Jalapeño” Inference Accelerator The companies detailed a custom inference processor developed for OpenAI’s expanding AI infrastructure, extending Broadcom’s role in hyperscale custom silicon. |
JUNE 9, 2026 AI Infrastructure | Broadcom Targets 20 GW of AI Compute Capacity with $35B Financing Broadcom outlined financing and infrastructure plans tied to large-scale deployment of custom accelerators, Ethernet fabrics and supporting AI systems. |
APRIL 14, 2026 Semiconductors | Broadcom Lands Major Meta AI Silicon Win With Multi-Generation MTIA Deal The multi-generation engagement expands Broadcom’s participation in Meta’s MTIA accelerator roadmap for AI inference and recommendation workloads. |
APRIL 6, 2026 Financials | Broadcom Secures Long-Term Role in Google’s TPU Roadmap Broadcom secured a long-term role supporting future Google TPU generations and networking components for next-generation AI rack infrastructure through 2031. |
