further evidence that people shoveling money into AI are either bad at math or blind to risk

@GaryMarcus Or great at recognizing geometrical progression and willing to accept the risk.

@GaryMarcus The valuation multiples of most stock market companies are scary right now. The level of PE ratios being traded at makes me very nervous

$SPCX: Well, even Jensen the Great Huang started having math problems, or caught some kind of virus when he traveled to China on AF-1 a few weeks ago.

AMZN is a 2,538x bagger. If SPCX were to do the same, its MCap would be $4,442 trillion, or 36x the current total world GDP. => SPCE's P/S = 36x if all world GDP is spent with SpaceX.

@GaryMarcus firms paying triple for compute just to shave 30 seconds off a support ticket is gonna age poorly

@GaryMarcus Or it could mean investors believe current returns don't fully reflect the long-term opportunity. Markets often price future outcomes before they appear in the data.

@GaryMarcus they've probably been asking AI to do the math.

@GaryMarcus ever seen a startup actually do the math before diving in?

@AGMacDougall @GaryMarcus @MediaTechDem He’s 100% correct

@GaryMarcus Any fear robs rationality.

FOMO perhaps most acutely.

(Just wait for the panic when ad profits melt away for some of them..)

@GaryMarcus 2.7x is a massive bar that almost nobody is saying out loud

kinda wild how much money is riding on a number nobody knows how to hit

@GaryMarcus the wharton math doesnt lie but the market doesnt care about math either

feels like we re already past the point of no return on this bet

@GaryMarcus Risk aversion gone extinct

@GaryMarcus the wharton paper didnt even factor in how fast these companies burn cash

2.7x seems optimistic

@GaryMarcus #3...they are just looking for the fast buck.

@lestercrafton what i call the trillion pound baby fallacy, after this facetious tweet