Full question:
By 2050, it will still not be possible to increase a healthy adult human’s IQ by at least two standard deviations in less than 30 days.
One of the questions from https://jacyanthis.com/big-questions.
Resolves according to my judgement of whether the criteria have been met, taking into account clarifications from @JacyAnthis, who made those predictions. (The goal is that they'd feel comfortable betting to their credance in this market, so I want the resolution criteria to match their intention.)
Related questions
@JimHays Hmm, is that going to be confusing to people? Maybe I should N/A and create one not phrased in the negative.
@IsaacKing You don’t have to on my account, but I do think it’s generally clearer to phrase questions in the positive
@Dreamingpast Presumably this is asking about the measure of IQ used in 2050. Unless Jacy says otherwise within the next few days, I'll codify that in the description.
@IsaacKing that'd be pretty weird. Metrics are to measure meaningful stuff in a frame of reference. We don't measure computer perf the same way as 30 years ago.
Someone with lots of mental math or iq test type questions practice will perform better but if the test in 2023 is measuring X while in 2050 is not measuring X but Y then it's going to be quite a useless comparison. We compare our standard deviations in the distribution of today, not that of 1900
@Dreamingpast the vocabulary iq is the only type that means anything. A vocabulary combined with a Turing test to use the word in conversation.
Like say I own a Mercedes, they have to know that car types exist and understand it has doors and is a sedan. That's about a thousand pieces of discrete information it has to remember.
If that costs 1mb to store then a million word vocabulary fills up the hard drive, then colloqiualisms and idioms explode that number which is why computers are dumb.