NewPhotonics and Tower Semiconductor Began Engine Shipments
The companies have launched volume shipments of laser-integrated optical engines for AI infrastructure.
Updated on Sept. 28, 2026 in Semiconductors
Live Poll
Do you believe rapid expansion in AI hardware infrastructure will improve the nation's economic productivity?
NewPhotonics and Tower Semiconductor have initiated high-volume shipments of their 800G and 1.6T photonic integrated circuits. These chips utilize the PH18DA silicon-photonics platform to support expanding AI infrastructure.
Why it matters
These integrated optical engines are essential for scaling the data transmission speeds required by modern artificial intelligence systems. By consolidating multiple optical components onto a single chip, the technology improves efficiency for high-capacity networking.
The circuits operate on the PH18DA silicon-photonics platform, which features heterogeneously integrated indium phosphide components. This architecture combines lasers, amplifiers, modulators, and photodetectors on a single chip.
The players
NewPhotonics
NewPhotonics is a technology developer specializing in photonic integrated circuits for high-speed data transmission.
Tower Semiconductor
Tower Semiconductor is a global specialty foundry that provides analog semiconductor solutions and manufacturing platforms.
The details
The collaboration leverages Tower's specialized silicon-photonics manufacturing process to pack high-performance optical functions into compact engines. This integration is designed to meet the rigorous data processing demands of large-scale AI environments.
Timeline
Volume shipments of 6.4T near-packaged optics chipsets are scheduled for the first half of 2027.
The Tech Race
This integration of indium phosphide components on silicon-photonics platforms represents a critical evolution in the race to overcome copper-based bandwidth limitations. It replaces legacy discrete optical systems with unified chipsets, positioning these firms as key suppliers in the AI infrastructure arms race.
The adoption of these high-performance optical engines will likely enable faster data processing speeds for AI-driven cloud services. Users may experience improved performance and reduced latency in AI-based applications as this hardware becomes standard in global data centers.
The takeaway
The move toward heterogeneously integrated optical engines signals that AI hardware is shifting toward deeper on-chip component consolidation. Organizations seeking to scale AI operations will increasingly rely on these high-density photonic platforms to maintain required transmission speeds.
What happens next
Volume shipments of 6.4T near-packaged optics chipsets are planned for the first half of 2027.
Further reading
Learn more about the latest innovations in Semiconductors.
Source note: This article includes information reported by Metal.
Live Poll
Do you believe rapid expansion in AI hardware infrastructure will improve the nation's economic productivity?
