Fully Homomorphic Encryption

Working Group Summary

Through the use and deployment of cryptographic libraries, specialist software toolchains and dedicated hardware and infrastructure, FHE can be leveraged to provide enhanced levels of protection for data in use. Supporting integration of FHE-enabled workflows with existing data practices will require technical resources and insight based on ongoing experimentation. The objectives of the working group are to investigate, understand and communicate the impact  of FHE in a cloud security context, and to establish best working practices and standards in the implementation and use of FHE as an information security tool.  What do we discuss? We discuss FHE in the context of cloud computing, with emphasis on practical elements of implementation, standardization and security modelling. The regulatory environment, impact on the control domains for cloud security, and establishment and evaluation of proofs-of-concept, also fall within the scope of the working group.

What is Fully Homomorphic Encryption?

Fully Homomorphic Encryption (FHE) is a cryptographic technology that enables computing over encrypted data. In traditional computing and cryptography, data can be protected by encryption when in storage or in transit over networks, but this protection must be stripped away before processing. This leaves data-in-use vulnerable to malicious attacks, accidental leakage, or unwarranted data sharing. FHE closes this vulnerability by providing a cryptographic system in which computing operations can be performed directly on the data without removing the protective encryption. 

Any computing operation can be performed under FHE, including more complex functionality such as machine learning and other data analytic processes. FHE also supports a range of different security models that allow for forms of secure multi-party computation and collaboration in which the inputs to the process remain totally private.

FHE offers significant improvements to the way in which access and analysis of sensitive data is handled in fields such as finance, healthcare, and government, where cryptographic assurances of protection throughout the data life-cycle are highly desirable. Access to sensitive information for processing purposes is critical to the modern world, yet satisfactory protection of this asset under existing models of information security is technically complex and difficult to achieve. By encrypting data throughout all stages of management and use, a wide range of challenges to information security including both internal and external threats can be better managed. 

In summary, FHE is a new and extremely powerful technology that extends mathematically assessable cryptographic security from storage and transit to the processing stage of data usage. As the support provided by advances in the core technology and infrastructure improves, FHE is likely to become a ubiquitous tool in data security, in much the same way as conventional cryptography has become universal.