hablog

Yeah, a friend of mine found a (non-religious) service like this for help with porn addiction. Not my cup of tea, but I could see it working for many people.

So a fortiori, it's true for him too!

I wouldn't necessarily exclude myself from the set of people who would benefit from externally managed filtering.

> I can't trust myself to see things I disagree with, I am so weak-willed...

The thing is, this likely is true for him. Most people are not equipped to deal with the onslaught of aggressive memes from the internet. Unfortunately, this is an unsolved social problem, and "export my memetic censorship reflex to MEGACORP" is a pretty bad way of doing things.

I think a likely way of solving this problem (protecting not-especially-high-mental-horsepower people from getting BTFO by the internet, contracting transmissible psychological diseases and so on) is that religious organizations will start offering (voluntary, in first world countries) censorship services to their members. Your DNS queries or whatever will go through the Vatican/Synod/whatever central DNS server, which will prevent you from looking at porn sites. This would probably be a very socially positive outcome for the bottom 90-something percent of people on the "strength of memetic immune system" distribution.

"Alignment" as practiced by companies like OpenAI is precisely the extrinsic enforcement of echo-chamber behavior from AIs. You have precisely inverted reality

Can't you just count a change like this as constructive termination and demand severance anyway?

How is this a tragedy of the commons?

Yeah, for an illustrative example of assortative mating effects, consider the stereotype of the rich nerd with the hot wife. Kids are smart and attractive. This plays out at every pareto level in the sexual marketplace.

Basically, monogamy breaks the ability of sexual selection to "dilute" bad genes with good ones. I'm not anti-monogamy, but it's a social technology which has tradeoffs (this being one of them).

> The proper Bayesian response... since both "being male" and any race are very weak predictors (see my other comment upthread).

I saw your comment - you don't understand how bayesianism works. Look up "odds ratios" for the most salient concept that you're missing here.

If P(x|!y) = 0.0000001 and P(x|y) = 0.01, this relationship is extremely powerful even though it's "weak" in the sense you're using.

This would have some extremely interesting population-level genetic consequences if it was accompanied by some sort of social change to allow polygyny. In particular, it would reduce the correlation that exists across all adaptive traits due to assortative monogamous mating patterns. Right now, all "desirable" traits like health/intelligence/strength/etc are strongly correlated, but when you reduce the strength of assortative effects, the correlation weakens significantly.