5

Assuming that cryptography falls into one of three areas:

  1. Awfully weak protocols and algorithms, that are trivial to break with a low amount of effort and resources

  2. Cryptographically strong protocols and algorithms that require more time than the lifetime of the Universe to actually crack

  3. The grey area of unproven, academically hard, or where anyone with a modest amount of resources can't break.

Question

  • Are there any weak, but imperfect homomorphic algorithms or approaches that are fast enough to be used in modern applications?

I'm not sure at what point weak cryptography is considered "obfuscation", but I hope that the relative investment in cryptography has a corresponding improvement in security of my plain text data.

6

There have been a few implementations of fully homomorphic encryption (2010). Fully homomorphic means it can perform addition AND multiplication (as opposed to partially homomorphic).

The answer to your question though relies on what you consider "weak" and sufficient for security. Does the encryption have to stand up for 100 years, or is it OK if it's cracked a week from now (short term secrets such as military troop movements)?

Anyway, homomorphic encryption really isn't ready for prime time. It's mostly a research area at the moment. The current "implementations" are very academic with little documentation and haven't been vetted (using encryption which hasn't been tested extensively is a fools errand).

Here's a link anyway if you want see what the current state of the art implementation looks like http://github.com/shaih/HElib/

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.