The Michael Anissimov vs. Dale Carrico debate (if one can ever label blog speechifying as such) and its parallel eruptions of comment groupthink have produced precisely one observation of lasting insight: Anne C’s suggestion that discussions of artificial intelligence begin from a proper understanding of the concept of “intelligence”:

For example, in many of the discussions of “intelligence enhancement” that I’ve come across, I’ve noticed an extremely dogmatic adherence to the notion of IQ as tested by common measures.

I’ve read plenty of papers on the subject myself, and the one thing that strikes me about much of the research (though some of the newer cog-sci stuff isn’t bad) is the tremendous amount of unexamined assumptions.

That is, you’ll see comments like, “well IQ correlates with Success In Life!” — where “success in life” is often defined purely in terms of being able to check off certain items on a narrowly-defined socioeconomic checklist.

Therefore, if you see the notion of “enhancing” humans so that they score better on common IQ tests to be an unquestionably good thing, you might want to think very carefully about what it means to “raise IQ”. In some respects, “raising IQ” almost by definition implies, “making people more in line with the things that current culturally and socially-influenced tests point to as valuable”. And that’s a subtle point, but an important one to understand (in my humble opinion).

Also, I’ve noticed that a lot of superintelligence-oriented folks seem to flip back and forth almost seamlessly between their own, private definitions of intelligence (e.g., that exemplified by people they think of as “smart”) and statistical constructs like g. That can get a bit confusing.

Not to be presumptuous or anything, but I would definitely like to see more acknowledgment from “intelligence enthusiasts” of the cultural, social, and historical factors that influence things like test scores.

Mind you, I’m not saying that everyone really has the exact same underlying ability set, and that all test score differences are the result of cultural bias — but I am saying that you can’t talk about intelligence as if it is a purely technical, and not a political, issue.

“Intelligence enthusiasts”, I like that. Too many extropian commentators share a strange attraction toward using general intelligence as a measure of fitness.