Lattica surveyed cryptographers, engineers, and researchers to understand where FHE is headed. The responses reveal a mix of optimism and skepticism: progress is happening, but challenges remain.
Here’s what the community thinks:
Lattica surveyed cryptographers, engineers, and researchers to understand where FHE is headed. The responses reveal a mix of optimism and skepticism: progress is happening, but challenges remain.
Here’s what the community thinks:
Most respondents are highly familiar with FHE, but that doesn’t mean it’s easy to use. Even experts highlight the difficulty of implementing FHE efficiently. Cryptographic tools exist, but practical deployment remains complex.
Our take: We believe FHE should be usable without deep cryptograpic expertise. Our work focuses on closing the gap between theoretical addvancements and real-world deloyment , making FHE accessible at scale.
Predictions varied:
Our take: Instead of chasing theoretical breakthroughs, we prioritize making FHE practical today by leveraging deep learning optimization methodologies, engineering best-practices and industry-leading tooling that enable fast development iterations.
Many respondents agreed that both hardware acceleration and better software optimizations are needed for FHE to scale.
We observe a growing number of hardware-software collaborations in the community, where startups and research groups are working together to optimize FHE performance.
Our take: We focus on building a software layer that fully utilizes specialized FHE hardware, ensuring that acceleration efforts translate into real-world performance gains. At the same time, our solutions work with today’s infrastructure while staying adaptable to future advancements.
FHE development today relies on a mix of open-source libraries like Open FHE, Concrete, SEAL, and Lattigo. While these libraries provide flexibility, they also contribute to a fragmented landscape where developers must carefully select schemes, parameters, and optimizations to fit their needs.
Our take: We believe FHE should be usable without deep cryptographic expertise. Our work focuses on closing the gap between theoretical advancements and real-world deployment, making FHE accessible at scale.
FHE is promising for privacy, but how does it fit into today’s regulatory landscape?
Our take: We believe that advancing the technical applicability of FHE is key to driving its standardization. Widespread adoption will require both practical implementations and clear regulatory frameworks, and these two must evolve together. Our focus is on making FHE technically viable at scale, which we see as a necessary step toward broader compliance and industry acceptance.
Many respondents see FHE as part of a broader privacy stack, often used alongside MPC, ZKPs, and Secure Enclaves. Some noted that FHE can reduce reliance on vendor-trusted hardware, while others pointed to hybrid models for balancing tradeoffs.
Our take: To enable more complex privacy-preserving architectures, FHE must first be made practical in specific, well-defined use cases. We focus on making FHE work in AI inference today, creating a foundation for broader adoption and future integrations with other privacy-enhancing technologies.
The survey shows growing confidence in FHE’s future. Yes, challenges remain, especially around performance, usability, and regulation, but innovation is accelerating.
For FHE to reach real-world adoption, both technical advancements and industry collaboration will be crucial. Practical implementations, hardware-software co-design, and clearer regulatory frameworks will all take part in that process.
Our take: FHE doesn’t have to be a distant vision. The key to adoption is making it work in specific, high-value applications today, while also building the foundations for broader adoption as the technology and ecosystem evolve.
Want to dig into the full survey data?
Let's Talk