I'd start by skimming some of the professor's papers. That will give you some intuition for what sorts of problems the professor will be able to give the best supervision on. Even if they leave things wide open for you, you're best off being supervised by someone who actually knows their stuff. If you're interested in something which they don't know anything about, you should look for another supervisor.
If looking over their past work doesn't give you any ideas about open problems they can help you with, then it's going to depend on what you already studied and are interested in. Blockchain? Electronic voting systems? Breaking or improving the security or performance of cryptographic primitives? Proving existence or impossibility of certain abstract properties of cryptosystems? Eg how does a paper like https://academic.oup.com/comjnl/article/49/3/310/564250 grab you?