Fully Homomorphic Encryption
In short: Fully homomorphic encryption (FHE) is a type of encryption scheme that allows arbitrary computation on encrypted data.
Fully homomorphic encryption (FHE) is a type of encryption scheme that allows arbitrary computation on encrypted data. The result stays encrypted, and once decrypted it matches what the same computation would have produced on the original data.
Why it matters
Normally, data has to be decrypted before it can be processed, which exposes it to whoever does the processing, such as a cloud provider. FHE removes that step. Earlier "partially homomorphic" schemes support only one kind of operation, such as addition or multiplication. A fully homomorphic scheme supports both, which is enough to build any computation. Cryptographer Craig Gentry published the first fully homomorphic scheme, and later research has made it much faster, but it is still far slower than computing on plain data. In blockchain, FHE is explored for confidential smart contracts, private voting and sealed-bid auctions, where a network must calculate a result without anyone seeing the individual inputs. It is related to, but different from, zero-knowledge proofs, which prove that a statement is true without revealing the data behind it.
Example
A user encrypts two numbers with their key and sends the encrypted values to a cloud service. The service adds them without ever decrypting them and returns an encrypted total. Only the user, who holds the secret key, can decrypt the result and read the sum. The service never learns either number or the answer.