> It’s roughly a 20,000x overhead. Would this be a realistic use case? A wants to send 256 bit message (M) to B. B sends already encrypted 256-bit AES key (

> It’s roughly a 20,000x overhead. Would this be a realistic use case? A wants to send 256 bit message (M) to B. B sends already encrypted 256-bit AES key (K) to A. A can use encrypted K to encrypt M and send it to B without knowing K. (essentially public key symmetric key)