AMD Acquires Enosemi to Advance Co-Packaged Optics for Artificial Intelligence Systems

AMD acquires Enosemi, expanding its expertise in photonics to fuel the next generation of Artificial Intelligence system interconnects.

AMD has officially acquired Enosemi, solidifying a strategic move aimed at accelerating innovation in co-packaged optics and photonics for next-generation Artificial Intelligence systems. Building on an established collaboration where Enosemi contributed as an external development partner, the acquisition brings Enosemi´s expertise and proven technology directly into AMD´s ecosystem. This move is expected to significantly enhance AMD´s capabilities to deliver high-performance photonics and interconnect solutions tailored to the growing demands of Artificial Intelligence workloads.

The Enosemi team, renowned for its elite experts and PhD-level talent, is headquartered in Silicon Valley and has demonstrated its ability to design, manufacture, and ship photonic integrated circuits at scale. This capability is described as a rare achievement in the industry, with only a select number of teams able to match such experience in photonic integration and volume production. Enosemi’s specialists are known for their technical rigor, deep engineering expertise, and a track record of reliable execution within the advanced photonics and semiconductor sectors.

By integrating Enosemi into its structure, AMD is positioning itself to more aggressively pursue high-performance interconnect solutions, a critical component for data-intensive Artificial Intelligence applications. The focus on co-packaged optics signals AMD´s intent to address bandwidth, latency, and energy challenges faced by future Artificial Intelligence hardware. This acquisition aligns with AMD´s rapidly evolving Artificial Intelligence strategy and strengthens its capacity to meet industry demands for faster, more efficient communication between processing components in both existing and emerging Artificial Intelligence systems.

73

Impact Score

Samsung to supply half of NVIDIA’s SOCAMM2 modules in 2026

Hankyng reports Samsung Electronics has secured a deal to supply half of NVIDIA’s SOCAMM2 modules in 2026 for the Vera Rubin Superchip, which pairs two ‘Rubin’ Artificial Intelligence GPUs with one ‘Vera’ CPU and moves from hardwired memory to DDR5 SOCAMM2 modules.

NVIDIA announces CUDA Tile in CUDA 13.1

CUDA 13.1 introduces CUDA Tile, a virtual instruction set for tile-based parallel programming that raises the programming abstraction above SIMT and abstracts tensor cores to support current and future tensor core architectures. The change targets workloads including Artificial Intelligence where tensors are a fundamental data type.

Contact Us

Got questions? Use the form to contact us.

Contact Form

Clicking next sends a verification code to your email. After verifying, you can enter your message.