Optalysys Raises US$31M to Push Silicon Photonics Into Commercial “Encrypted Compute” Hardware

Key takeaways

• • Optalysys raised $31M in a Series A extension, led by Northern Gritstone, with participation from imec.xpand, Lingotto Horizon, and the UK government’s NSSIF.
  • The company says it will use the funds to accelerate commercialization of its proprietary photonic chips and expand into the U.S.
  • Optalysys is positioning photonic computing as a path to high-performance, energy-efficient processing for AI/cloud and as a hardware accelerator for privacy-preserving computation, especially Fully Homomorphic Encryption.

On January 22, 2026, Leeds-based Optalysys announced it raised £23 million in a Series A extension round. The company named Northern Gritstone as the lead investor, with participation from imec.xpand, Lingotto Horizon, and the UK government’s National Security Strategic Investment Fund (NSSIF).

Optalysys says the capital will go toward commercializing its photonic chips and expanding into the U.S. NSSIF frames the company’s mission building an “encrypted compute system powered by light,” enabling data to be processed while remaining encrypted end-to-end.

AI becoming a data-movement and privacy bound

AI infrastructure isn’t only gated by raw compute anymore. Power and grid capacity are now a primary constraint for data-center buildouts, 72% of respondents in a Deloitte survey called it very or extremely challenging.

At the same time, the “memory wall” problem is getting harder as accelerators can’t deliver full value if data can’t be fed fast enough (bandwidth and throughput constraints increasingly cap real performance).

Layer privacy on top, especially for regulated industries, and you get a market pull for architectures that don’t just compute faster, but can compute securely and efficiently.

What Optalysys says is different about its silicon photonics approach

Optalysys’ standout narrative is FHE (Fully Homomorphic Encryption)—a method that allows computation on encrypted data without decrypting it during processing. That’s powerful, but historically too slow and costly for many real deployments.

Optalysys argues that photonic processing can reduce the overhead that has held FHE back—and it has been building a software ecosystem story around that claim. For example, the company partnered with Google HEIR to integrate Optalysys’ photonic processing technology into HEIR’s compiler toolchain for FHE workloads.

It also announced a partnership with Duality Technologies aimed at accelerating FHE adoption by connecting Optalysys’ tech with privacy-preserving software stacks and the broader FHE ecosystem.

Optalysys positions its platform as a blend of silicon photonics + digital processing, aiming to improve energy efficiency and reduce bottlenecks tied to moving data electrically at scale.

The company also points to early productization efforts (e.g., server-class systems for encrypted workloads), but the commercial question is how quickly those concepts translate into repeatable deployments and measurable benchmarks.

CEO Dr. Nick New emphasizes this point: “We are at a defining moment in the evolution of computing. Photonic computing opens up fundamentally new capabilities, allowing data to be moved and processed with far greater speed and efficiency.”

Industry research supports this perspective: optical accelerators that leverage light can dramatically speed up key FHE operations like Fourier Transforms, which are central to many homomorphic encryption algorithms.

Optalysys has been building credibility well beyond funding headlines. Notable collaborations include:

• Google HEIR partnership: Optalysys has integrated its photonic tech with Google HEIR, a compiler for FHE workloads, demonstrating substantial performance improvements in encrypted computation trials.
• Duality Technologies collaboration: A partnership with Duality aims to accelerate quantum-resistant encryption use cases by combining Optalysys’ cloud acceleration tech with privacy-preserving software stacks.

These initiatives showcase how photonics is being integrated into both software ecosystems and hardware roadmaps, not just kept as isolated lab projects.