"All you need to do is make the expected value of an attack against the system worth less than just doing some actual research." you'll always have some people

"All you need to do is make the expected value of an attack against the system worth less than just doing some actual research." you'll always have some people willing to take a gamble on the risk to get the reward, especially if it's a low % chance that they get any penalty at all. on top of the conscientious choice, humans are bad at estimating low-likelihood risks.