Why are these people obsessed with a measurement of people that has the worst statistical validity known to mankind? IIRC Taleb ripped IQ-tards to shit a couple of years ago with an extended rant / published paper on the various flaws, but even from a non-stats nerd perspective the idea that you can infer the “IQ” of an entire country from the studies that these people quote is absurd. We’re talking “I measured the IQ of a bunch of kids in an orphanage in 1968, so obviously I can infer from this the value of the entire nation” levels of absurd here. Even taking their words at face value (lol) the entire endeavour is ludicrous.
The whole thing just seems completely pointless: what’s the end goal here? Being able to line people up like trading cards achieves what exactly?
(Which leads inevitably to the depressing reality that the “why” is obvious, but unspoken of course: They just don’t want to come out and say it.)
So it looks like Mr. “Not consistently candid” has been at it again?
I will admit that they got me with this one: I genuinely thought the FrontierMath results meant something real. I didn’t think they would be that brazen about rigging a benchmark that was explicitly advertised as being kept private so that AI companies couldn’t train on the questions. More fool me I guess.