the problem lies with the fact that IQ scores are not statistically translatable between tests, and most tests have no actual standards for intelligence.
the common problem with almost all IQ tests is that they exist to boost the ego, ie they are "feel good" tests.
unless you specifically go out of your way to take a test designed to measure IQ in the 160 range, chances are you took a feel-good online test.
people score 140s on online tests and think that actually means they have genius or near genius level of intelligence, but they dont consider what those numbers were actually being compared to.
how many people who have taken "IQ tests" have actually done so in a proctored environment set by a legitimate genius-tier IQ society (mega, mensa, prometheus etc)?
probably less than 1%, i'd wager.
therein lies the problem. IQ scores do not translate between tests. an IQ of 140 on iqtest.com does not equal an IQ of 140 on a true proctored test.
and then how many people actually took a mensa-level test or greater?
a few thousand?
does a few thousand accurately represent 7 billion people?
does a 140 IQ on a test measured from 10,000 people mean you have a genius level IQ compared to 7,400,000,000?
if you think so, you might want to reconsider that "140 IQ"
and what about the tests themselves? the hardest IQ tests arent even timed, you're given upwards of a month to solve them. but online tests are usually 15-25 minutes. and the material? usually 140 IQ tier tests are just matrices, but the harder IQ society tests are all word problems with virtually no pattern recognition involved.
there's really no actual coordination or consistency involved in IQ testing, so there is no reason to assume the tests are good for anything other than an ego boost unless you specifically take a "real" IQ test to join a real genius tier IQ society