Computing: Why Apple's homomorphic encryption is a big deal

Excerpt:

However, for years [homomorphic encryption] was far too slow and resource intensive for all but a few very niche use cases, with simple operations sometimes taking hours or even days to complete.

But that's changing now, thanks to advances in the underlying technology and efficiencies relating to its use, according to Ellison Anne Williams, CEO of privacy-enhancing technology company Enveil. 

Enveil uses Microsoft's SEAL library in its ZeroReveal Compute Fabric Server, which is is designed to process encrypted queries over data without decrypting the queries or revealing their content.

The company implements privacy-enhancing technologies (PETs) including HE in companies including large financial firms, enterprises and public sector organisations, and there are more and more places where it's a good fit. 

"HE can now operate at the speed of business and mission for an increasing number of use cases," Williams told Computing. "Encrypted searches can now be performed over millions of data records and returned within seconds - rather than days."

Nevertheless, decrypting the data prior to analysis is still significantly quicker than using HE, so it won't be replacing standard cryptosystems for most processes. Instead it's about enabling new use cases, she said.   

"It's bringing entirely new capabilities to the table by allowing organisations to leverage data across boundaries while respecting regulatory and other security requirements. For entities where that capability hasn't before been achievable, adding seconds to a workflow is an easy trade-off."