Cryptographic Hash Functions

Background

Have you ever considered how we verify the authenticity of certain messages? We use a One Way Function called Hashing. Basically, we take anything, and convert it to a seemingly random (but entirely deterministic) string.

There’s also the desire for a collision resistant function. What does that mean? A function that makes it really hard to find two messages $m_1$ and $m_2$ such that $f(m_1)=f(m_2)$.

For what?

It’s used for a lot:

• File integrity when downloading something (you can compare your hash to the sites)
• Password verification - instead of storing plaintext passwords - you store their hashes.