In a press release issued on 8 October 2025 from its Sunnyvale campus, NTT Research, Inc. announced that its Physics and Informatics (PHI) Lab, in partnership with Cornell University and Stanford University, has created the world’s first programmable nonlinear photonic waveguide. The device can be reconfigured in real-time to perform any of several nonlinear-optical functions, such as second-harmonic generation, arbitrary pulse shaping, or holographic light synthesis, by projecting a structured light pattern onto a silicon-nitride core. The breakthrough, detailed in a paper published online in Nature and slated for print on 13 November, promises to break the long‑standing “one device, one function” rule that has constrained photonic engineering for decades.
Why NTT’s Waveguide Breaks the One-Device-One-Function Rule
Traditional photonic chips are built to perform a single, hard‑wired operation. A waveguide that doubles a laser’s frequency must be fabricated with precisely engineered dimensions; a waveguide that shapes pulses must have a different geometry. The cost of producing separate components for each task, combined with the sensitivity of optical fabrication to microscopic defects, has limited the scalability of photonic systems. NTT’s new waveguide sidesteps these constraints by making the core’s nonlinearity itself a variable. Using an external “programming light,” the device’s silicon‑nitride core develops spatially patterned refractive indices that dictate the type of nonlinear interaction that occurs when a signal light passes through. By simply changing the pattern of the light that illuminates the chip, the same physical structure can instantaneously switch between at least four distinct functions.
“These results mark a departure from the conventional paradigm of nonlinear optics, where device functions are permanently fixed during fabrications,” said Ryotatsu Yanagimoto, a scientist at NTT Research. , Ryotatsu Yanagimoto, NTT Research
The technology leverages electric‑field‑induced nonlinearities, a relatively unexplored effect, to create a dynamic refractive‑index landscape. In experiments, the team demonstrated widely tunable second‑harmonic generation spanning a 100‑nanometre bandwidth, arbitrary pulse shaping over a 10‑terahertz range, and the holographic generation of spatio‑spectrally structured light with a fidelity exceeding 95 %. Crucially, the system tolerates fabrication imperfections and environmental drifts, because the re‑programmable pattern can be adjusted on the fly to compensate for any variations.
How Cornell and Stanford Collaborated on Reconfigurable Photonics
The joint effort drew on complementary strengths from the two universities. Cornell’s Peter L. McMahon, an associate professor of electrical engineering, contributed expertise in silicon‑photonic design and waveguide fabrication. Stanford’s involvement centred on developing the structured‑light projection system that writes the nonlinear patterns onto the chip. The collaboration was coordinated through NTT’s PHI Lab, which houses interdisciplinary teams focused on quantum information, neuroscience, and photonics.
McMahon explained the design philosophy: “We wanted to create a platform that could serve as a universal building block for optical circuits,” he said. The result is a waveguide that can be addressed by a computer‑controlled light‑shaping apparatus, enabling rapid prototyping of optical functions without the need for new masks or lithography steps. The project’s success demonstrates how academia and industry can combine resources to push photonic integration beyond its conventional limits.
The joint research also highlighted the importance of inverse design techniques. By feeding a target optical response into a machine‑learning algorithm, the team could generate the exact light pattern needed to produce that response in the waveguide. This closed‑loop approach allows designers to specify arbitrary functions, such as generating a particular quantum state of light, and have the chip realise it without manual tweaking.
The Path Forward for Large-Scale Optical Circuits
NTT’s announcement comes at a time when the photonic‑integrated‑circuit market is projected to surpass $50 billion by 2035, according to IDTechEx. The new programmable waveguide could accelerate this growth by reducing the number of discrete components required in a system. In telecommunications, for instance, a single chip could replace separate modulators, frequency converters, and pulse shapers, simplifying the architecture of 5G and forthcoming 6G networks. The ability to tune light sources across a wide spectral range also benefits optical communication systems that rely on wavelength‑division multiplexing.
In quantum computing, programmable quantum frequency converters and quantum light sources could enable more flexible architectures. A chip that can re‑configure its nonlinear response on demand would allow quantum processors to adapt their gate sets without hardware changes, a critical capability for scaling up qubit numbers while maintaining coherence. Moreover, the technology could improve quantum networking by enabling on‑demand conversion of entangled photons between different wavelengths, facilitating long‑distance quantum key distribution.
Beyond communications, the platform offers advantages for sensing and imaging. Programmable structured light sources can adapt to changing sample conditions, improving resolution and contrast in microscopy or enabling adaptive illumination in LiDAR systems. In industrial manufacturing, real‑time reconfigurability could allow a single photonic chip to perform a suite of tasks, from laser cutting to surface inspection, reducing equipment footprint and cost.
Future research will explore extending the device’s capabilities to genuinely quantum regimes. While the current demonstrations focus on classical nonlinear optics, the underlying principle, controlling the refractive index landscape with light, could be harnessed to shape quantum states of light in situ. Additionally, the team plans to investigate other materials that exhibit stronger electric‑field‑induced nonlinearities, potentially widening the operational bandwidth and reducing the power required for re‑programming.
Yanagimoto’s Vision for Quantum Frequency Conversion
Ryotatsu Yanagimoto has long envisioned photonic devices that can be re‑configured on demand to meet the needs of emerging quantum technologies. In a blog post accompanying the Nature paper, he outlined the broader implications of programmable nonlinear photonics: “For the first time, a path forward has been created to apply nonlinear optics to large‑scale optical circuits, reconfigurable quantum frequency conversion, arbitrary optical waveform synthesizers and widely tunable classical and quantum light sources, all of which are vital to enabling advanced computing and communications infrastructure.”
Yanagimoto emphasises that the ability to change a device’s function without physical alteration is not merely convenient; it is essential for next‑generation systems that must adapt to varying workloads, error rates, and environmental conditions. By embedding programmability into the very heart of the photonic platform, NTT’s waveguide paves the way for modular, upgradable optical architectures that can evolve alongside software and algorithmic advances.
The PHI Lab’s broader mission, to study computation within the fundamental principles of quantum physics and brain science, finds a natural ally in this technology. As quantum processors grow in complexity, the need for flexible, low‑loss photonic interconnects will become ever more pressing. A programmable waveguide that can act as a tunable quantum frequency converter, a pulse shaper, or a holographic projector could become a cornerstone of the quantum hardware stack.
In short, NTT’s programmable nonlinear photonic waveguide represents a paradigm shift. By dissolving the rigid one‑device‑one‑function model, it unlocks a new design space in which optical circuits can be written, rewritten, and re‑used at the speed of light. The implications ripple across telecommunications, quantum computing, sensing, and beyond, promising a future where photonic systems are as adaptable as their electronic counterparts.
