My uneducated guess is that IQ is partially driven by something kind of silly and basic like just sheer speed or how well your neurons can signal to each other or uptake transmitters or how many of a specific kind of structure you have. Something that could increase a persons grey matter, connection between different areas, growth of temporal or parietal or frontal lobes, just adds more artificial cortical columns or pyramidal neurons I think would allow for the brain to have more resources to start thinking about things better. I honestly think the 30 days is the part that makes this more unrealistic instead of the two deviations. If you lets someone brain grow the structures needed for higher iq, I imagine it would still require some kind of training and use in order for it to be developed and structured in such a way as to translate to more IQ. If you remade the question be exactly the same but with a two to five year time period for the growth to happen I think it becomes a lot more realistic (& would honestly still be extremely impactful imo)

Does it require the treatment being worth >=2 std on 100% of people who haven't had it? Or would increasing any adult's IQ by that much count?

What about treatments that work on 10% of the population? And 1%

Will probably resolve the same as Metaculus

https://www.metaculus.com/questions/1625/by-2050-technique-raises-iq-by-30-pts/

See also the negated version of this question:

(It's the exact same question, I'm just curious which title phrasing gets more interest.)

@IsaacKing Arbed it. I wonder if there are some tail risks I'm missing?