Samsung Electronics introduced Wednesday its latest advancements in foundry technologies and vision for the AI era at the Samsung Foundry Forum (SFF) 2024 in San Jose, California.

The event, themed "Empowering the AI Revolution," showcased Samsung's enhanced process technology roadmap, which featured two new nodes: SF2Z and SF4U. 

SKOREA-ECONOMY-SAMSUNG
A woman walks past the Samsung logo displayed on a glass window at the company's Seocho building in Seoul on April 30, 2024.
(Photo : JUNG YEON-JE/AFP via Getty Images)

Samsung Unveils its AI Innovations, Foundry Technologies

Dr. Siyoung Choi, President of Samsung Electronics Foundry Business, emphasized the pivotal role of advanced semiconductor solutions in enabling AI innovations. 

He highlighted Samsung's proprietary gate-all-around (GAA) technology, optimized for AI chips and designed to integrate co-packaged optics (CPO) for rapid, low-power data processing capabilities. This integration aims to provide comprehensive AI solutions tailored to meet evolving market demands.

The forum included presentations from industry leaders such as Arm CEO Rene Haas and Groq CEO Jonathan Ross, highlighting Samsung's collaborative efforts to address emerging AI challenges. Thirty partner companies participated in the event.

Samsung unveiled SF2Z, its latest 2nm process featuring optimized backside power delivery network (BSPDN) technology. This innovation enhances power efficiency and performance for high-performance computing (HPC) applications, significantly reducing voltage drop compared to previous generations.

Mass production of SF2Z is scheduled for 2027, promising advancements in semiconductor manufacturing. SF4U, a high-value 4nm variant leveraging optical shrink technology, is set for mass production in 2025.

Samsung also reported that its preparations for SF1.4 (1.4nm) are going well, and it targets mass production by 2027. The company is also conducting research on sub-1.4nm process advancements through innovative materials and structures.

With the onset of the AI era, gate-all-around (GAA) technology remains pivotal in Samsung's strategy to meet AI's power and performance requirements. Samsung plans to extend GAA to its upcoming 2nm process as it hits its third year of mass production.

Read Also: [LEAK] Samsung Galaxy Buds 3 Pro Appears to Have Noticeable Shape Similar to Apple AirPods

Samsung AI Solutions

Samsung AI Solutions was also unveiled at the forum. It integrates capabilities across Foundry, Memory, and Advanced Package (AVP) businesses to offer tailored solutions for AI applications.

This collaborative approach optimizes supply chain management (SCM), resulting in a remarkable 20% reduction in total turnaround time (TAT). Samsung's vision includes introducing a comprehensive CPO-integrated AI solution by 2027, which will simplify AI implementation for customers.

Following SFF 2024, Samsung will host the annual Samsung Advanced Foundry Ecosystem (SAFE) Forum with the theme "AI: Exploring Possibilities and Future."

Scheduled for June 13, this forum will convene industry leaders, including Siemens CEO Mike Ellow, AMD Vice President Bill En, and Celestial AI CEO David Lazovsky. Discussions will focus on customizable technologies and collaborative initiatives tailored to advance chip and system design technologies.

Additionally, the Multi-Die Integration (MDI) Alliance workshop will follow, building on last year's launch to explore 2.5D and 3D IC designs. 

"At a time when numerous technologies are evolving around AI, the key to its implementation lies in high-performance, low-power semiconductors," Choi said in a press release.

"Alongside our proven GAA process optimized for AI chips, we plan to introduce integrated, co-packaged optics (CPO) technology for high-speed, low-power data processing, providing our customers with the one-stop AI solutions they need to thrive in this transformative era," he added.

Related Article: Samsung Galaxy AI's Live Translate Expands to Non-Samsung Apps, New Features Coming for Foldables

Byline

ⓒ 2024 TECHTIMES.com All rights reserved. Do not reproduce without permission.
Join the Discussion