Talk:Security question

database match failures

This article needs major improvement.

There are two very different kinds of "security questions" in common use now.

One kind, which the article currently is mostly about, uses "answers" that have been supplied previously by the user/customer/client, which may or may not be "true" or important person factoids -- the answers might be arbitrary or even nonsense. These are just another form of password that require repeating at some future time data supplied earlier.

But there is a very different kind of "security question" also in common use now, which requires the user/customer/client to "prove" their identity by furnishing answers to obscure personal questions about their past which match data held in various databases. LexisNexis is a common source of this "service" to many institutions. It is common for consumers to "fail" this test either because the questions are too obscure or because some of these vast hidden databases are garbled/wrong. Consumers are unaware of these data sources, do not have access to their contents, and are mostly powerless to correct the wrong data about themselves. -71.174.180.38 (talk) 15:23, 13 October 2016 (UTC)