I think a big part of it comes down to what threats are there in theory and what threats are there actually. The problem is that the theoretical threats are possible, they're not unrealistic and that's why it doesn't feel good to not be protected against the theoretical threats but we maybe need to try and accept they are too unlikely to be active threats. Trying to protect from theoretical threats is kind of like trying to protect your house from having an airplane fall down from the sky into your house. Or maybe this is just my trying to cope.

And how do we know what threats are theoretical vs active threats? Just have to keep learning and learning, it takes a long time. Talking in privacy and security communities can help speed up the learning.

We've had meetings spelling out to users what they should look for in a suspicious email. Then, once a week we would send out an email that was either legitimate or suspicious. We would ask them to look closely at the email and mark down on the questionnaire whether the email was suspicious or legitimate. A not insignificant number of people failed the test every week. Your average user just isn't equipped with the mindset they need to be safe on the internet.