4 ms·
Wait... Given enough SSNs as training data, could someone make an ML model that churns our mostly valid SSNs? Because that could be really really bad
by cgb223 7y ago
Wait...
Given enough SSNs as training data, could someone make an ML model that churns our mostly valid SSNs?
Because that could be really really bad
- JumpCrisscross 7y ago> Given enough SSNs as training data, could someone make an ML model that churns our mostly valid SSNs? Most SSNs are deterministically derived. Given a single valid SSN, you can often generate millions of valid preceding values.
- ageitgey 7y agoThat changed in 2011. Now they are random. But it is still true for older numbers. https://www.ssa.gov/employer/randomization.html https://www.ssa.gov/employer/randomization.html
- pliny 7y agoDon't you need at least one other piece of identifying info to make that useful? Like, I could churn out mostly valid gmail passwords all day, but I have no idea which users have those passwords.
- jakelazaroff 7y agoNo need to get that involved: for (let i = 0; i < 1000000000; i++) console.log(i); Every single valid social security number, and several (e.g. 000-00-0000) that probably aren't. You can store the whole thing in about a gig.
- throwawaywego 7y agoWorse. You can make the ML model spit out the SSNs it was trained on. That's a problem when you can't manually curate billions of documents. If you didn't look, you wouldn't even know they were there.