Deep Tech, High Performance Computing

NewPhotonics and Tower Semiconductor Ramp Up Optical Engine Production

Avatar of Ivy Delaney
September 18, 2026 by Ivy Delaney

Tower Semiconductor and NewPhotonics have begun high-volume shipments of photonic integrated circuit optical engines, delivering 800G to 1.6T data transmission capacity for artificial intelligence infrastructure. The companies are commercializing a manufacturing process for these chips using Tower’s PH18DA platform, which enables monolithic integration of lasers, modulators, and detectors onto a single chip. “The next phase of AI infrastructure demands optical connectivity that scales in both production and performance,” said Doron Tal, SVP and GM of NewPhotonics, adding that the partnership is translating laser-integrated photonics into a practical, high-volume choice for customers.

Laser-Integrated PICs Enable 800G to 1.6T AI Interconnect Shipments

High-volume shipments of NewPhotonics’ NPG102 PIC products now support data rates from 800G to 1.6T, directly addressing the escalating bandwidth demands of artificial intelligence infrastructure. These chips feature 200G-per-lane designs and integrate lasers directly onto a monolithic photonic integrated circuit, a key advancement in optical signal processing. This unified architecture delivers improved power efficiency, bandwidth density, and manufacturing consistency for original equipment manufacturers and data center operators.

The availability of these PICs represents a shift from development to practical deployment, offering a scalable solution for AI interconnects. Tower Semiconductor’s PH18DA silicon photonics platform is central to this production ramp, enabling the heterogeneous integration of indium phosphide components alongside silicon. This allows for monolithic integration of lasers, semiconductor optical amplifiers, modulators, and photodetectors on a single chip, reducing complexity and variability in the manufacturing process.

According to Dr. Ed Preisler, Vice President and General Manager of Tower Semiconductor’s RF Business Unit, “We are excited to see our vision of integrating lasers with high-performance photonic integrated circuits come to fruition through scale-ready solutions optimized for FRO/LRO and NPO architectures.” Tower’s expertise in high-value analog semiconductor solutions, combined with NewPhotonics’ optical chip design, has resulted in a commercially viable pathway for laser-integrated photonics. The collaboration between the two companies extends beyond current shipments, with plans for 6.4T Near-Packaged Optics (NPO) solutions slated for volume production in the first half of 2027.

These future chipsets, designated NPC505 and compliant with the CPX-MSA standard, indicate a clear roadmap for significantly higher bandwidths. NewPhotonics is positioned to capitalize on this trajectory, having been the first to bring heterogenous laser-integrated optical engines on the PH18DA platform into high-volume production.

Tower Semiconductor, founded in 1993 and headquartered in Migdal Haemek, Israel, has increasingly focused on silicon photonics as a core competency. The company’s high-volume silicon photonics platform not only supports NewPhotonics but also is the foundation for a high-volume partnership with Xanadu, focused on manufacturing photonic quantum chips for fault-tolerant quantum computing. This diversification into quantum hardware demonstrates Tower’s commitment to advanced technologies beyond traditional optical communications. A recent partnership with OpenLight, announced on August 11, 2026, integrates design tools into Tower’s indium phosphide platform, accelerating photonic chip design processes.

More like this
Artificial IntelligenceAMD EPYC chips now handle every step of agentic AI workflows
PhysicsNew magnets aim to meet positron demands of 91km collider
Artificial IntelligenceQuantum materials could cut AI data centre energy use
Deep TechMIT builds a robot lab for optics experiments, cutting setup time
Stay current

See today’s quantum computing news on Quantum Zeitgeist for the latest breakthroughs in qubits, hardware, algorithms, and industry deals.

Tags:
Silicon Photonics
Avatar of Ivy Delaney

Ivy Delaney

Ivy Delaney has been working with neural networks and machine learning since the mid-nineties, back when a couple of hidden layers and a long afternoon of training counted as ambitious. She has watched the field go from academic curiosity to the thing quietly running underneath everything, and she brings that long view to quantum computing. For Quantum Zeitgeist she covers the ground where the two fields meet. That means quantum machine learning and the variational algorithms it leans on, and it also means the less glamorous but more interesting story of classical machine learning already doing real work inside quantum machines, decoding error-correcting codes, calibrating noisy hardware and learning the error models that simulators depend on. She writes about the hardware those algorithms have to run on too, and about the post-quantum cryptography scramble that the same hardware has set off. Her stories typically start with the paper, whether that is peer-reviewed work, conference proceedings or an arXiv preprint, with the source linked so you can hold a claim up against the research it came from. She is unimpressed by benchmarks that will not say what they beat, and by demonstrations that only work in the press release.

Latest Posts by Ivy Delaney: