Hello.

Ilya Kaminsky, “A City Like a Guillotine Shivers on Its Way to the Neck”, Deaf Republic

The whole point of God is that he’s omnipotent and omniscient, right? He can be blamed for doing nothing because he was capable of doing something. The rest of us have valid excuses. He doesn’t.

preamble

Sometimes I wonder what the MIRI-type crowd thinks about some issue related to their interests.  So I go to alignmentforum.org, and quickly get in over my head, lost in a labyrinth of issues I only half understand.

I can never tell whether they’ve never thought about the things I’m thinking about, or whether they sped past them years ago.  They do seem very smart, that’s for sure.

But if they have terms for what I’m thinking of, I lack the ability to find those terms among the twists of their mirrored hallways.  So I go to tumblr.com, and just start typing.

parable (1/3)

You’re an “agent” trying to take good actions over time in a physical environment under resource constraints.  You know, the usual.

You currently spend a lot of resources doing a particular computation involved in your decision procedure.  Your best known algorithm for it is O(N^n) for some n.

You’ve worked on the design of decision algorithms before, and you think this could perhaps be improved.  But to find it, you’d have to shift resources some away from running the algorithm for a time, putting them into decision algorithm design instead.

You do this.  Almost immediately, you discover an O(N^(n-1)) algorithm.  Given the large N you face, this will dramatically improve all your future decisions.

Clearly (…“clearly”?), the choice to invest more in algorithm design was a good one.

Could you have anticipated this beforehand?  Could you have acted on that knowledge?

parable (2/3)

Oh, you’re so very clever!  By now you’ve realized you need, above and beyond your regular decision procedure to guide your actions in the outside world, a “meta-decision-procedure” to guide your own decision-procedure-improvement efforts.

Your meta-decision-procedure does require its own resource overhead, but in exchange it tells you when and where to spend resources on R&D.  All your algorithms are faster now.  Your decisions are better, their guiding approximations less lossy.

All this, from a meta-decision-procedure that’s only a first draft.  You frown over the resource overhead it charges, and wonder whether it could be improved.

You try shifting some resources away from “regular decision procedure design” into “meta-decision-procedure-design.”  Almost immediately, you come up with a faster and better procedure.

Could you have anticipated this beforehand?  Could you have acted on that knowledge?

parable (3/3)

Oh, you’re so very clever!  By now you’ve realized you need, above and beyond your meta-meta-meta-decision-procedure, a “meta-meta-meta-meta-decision-procedure” to guide your meta-meta-meta-decision-procedure-improvement efforts.

Way down on the object level, you have not moved for a very long time, except to occasionally update your meta-meta-meta-meta-rationality blog.

Way down on the object level, a dumb and fast predator eats you.

Could you have anticipated this beforehand?  Could you have acted on that knowledge?

But by the same token, it’s really dumb to assume most people who talk about open borders are doing anything besides holding a directional goal if they’re talking about current policy prescriptions at all! Like, even ignoring the fact that we’re dealing with a bunch of libertarian-neighborhood folks who have a long track record of simultaneously a) describing far-off political systems that require a lot of interlocking pieces to work properly even if you believe their analysis and b) not seriously trying to foment revolution and, to the extent they engage at all in the political process, mostly working for incremental change, the fact that in this case there is no possible concrete path toward the feared outcome should be a sign that the people involved are not actually suggesting it as a standalone immediate policy proposal?

But by the same token, it’s really dumb to assume most people who talk about open borders are doing anything besides holding a directional goal if they’re talking about current policy prescriptions at all! Like, even ignoring the fact that we’re dealing with a bunch of libertarian-neighborhood folks who have a long track record of simultaneously a) describing far-off political systems that require a lot of interlocking pieces to work properly even if you believe their analysis and b) not seriously trying to foment revolution and, to the extent they engage at all in the political process, mostly working for incremental change, the fact that in this case there is no possible concrete path toward the feared outcome should be a sign that the people involved are not actually suggesting it as a standalone immediate policy proposal?

and apparently, when she straight-faced, lifted her glass of chardonnay to her ear and swirled it, she baffled the other winery-goers so much that they all instinctively did exactly the same thing. that sort of crowd-think, that you aren’t quite sure what’s right but you’ll be damned if you’re mistaken for the plebian that you are

but watching that was enough for my mom to lose her composure, laugh so hard she cried, and spill the entire glass all over herself, in the process getting kicked out of the winery

it turns out how much I like you is basically uncorrelated with your Objective Goodness

[I]t is sometimes possible with substantial effort to say that policy X has been checkmated, at least in terms of consequentialist supporting arguments. Yes, many social policy chess boards are big, and so it can take a lot of time and expertise to check all the moves. But sometimes a person has done that checking on policy X, and then frequently encounters others who have not so checked. Many of these others will defend X, basically randomly sampling from the many failed arguments that have been offered so far. […]

As an example, let me offer blackmail. I’ve posted here seven times on this blog on the topic, and in one of my posts I review twenty related papers that I’d read. I’ve argued many times with people on the topic, and I consistently hear them repeat the same arguments, which all fail. So I’ll defend the claim that not only don’t we have good strong consequential arguments against blackmail, but that this fact can be clearly demonstrated to smart reasonable people willing to walk through all the previously offered arguments.

To review and clarify, blackmail is a threat that you might gossip about someone on a particular topic, if they don’t do something else you want. The usual context is that you are allowed to gossip or not on this topic, and if you just mention that you know something, they are allowed to offer to compensate you to keep quiet, and you are allowed to accept that offer. You just can’t be the person who makes the first offer. In almost all other cases where you are allowed to do or not do something, at your discretion, you are allowed to make and accept offers that compensate you for one of these choices; blackmail is a puzzling exception to this general rule.

Most ancient societies simply banned salacious gossip against elites, but modern societies have deviated and allowed gossip. People today already have substantial incentives to learn embarrassing secrets about associates, in order to gain social rewards from gossiping about those to others. Most people suffer substantial harm such gossip, and it makes them wary about who they let get close to them, and induces them to conform more to social pressures regarding acceptable behaviors.

For most people, the main effect of allowing blackmail is to mildly increase the incentives to learn embarrassing secrets, and to not behave in ways that result in such secrets. This small effect makes it pretty hard to argue that for gossip incentives the social gains out weigh the losses, but for the slightly stronger blackmail incentives, the losses out weight the gains. However, for elites these increases are far stronger, making elite dislike plausibly the main consequentialist force pushing to keep blackmail illegal.

I agree with Hanson. I don’t think blackmail should be illegal, nor do I think in most cases it is even morally wrong. (I think it’s morally wrong in the context of blackmailing people where the consideration is not exposing their own criminal activity, which can be handled by whistleblower protections.)

I want to specifically address the privacy issue, which Hanson addresses elsewhere. The concern is that people might be blackmailed over things like naked photos of them or videos of them having “vanilla” sex with their spouse. These are embarrassing not because they show illegal or immoral activity, but because it’s simply an invasion of their privacy.

But in the case where the photos are obtained illegally, there is no need for a special law here. It’s already illegal to do the underlying thing (post them online for free), so of course it should be illegal to do it for money! And to deal with matters like “revenge porn”, the law can just have a default presumption that one does not consent to allow one’s partner or former partner to upload such photos they may possess without one’s permission. They’re issued under restrictive title.

My idea of blackmail is something, like: a woman (possibly hired by an evil party) flirts with a male politician; this leads to sexy things; the woman takes a selfie, and threatens to publish the selfie unless the politician gives the evil party some benefit.

At no point is any of the photos obtained illegally. But clearly it is bad for society for the politician to be acting in the interests of the evil party, rather than in the interests of voters.

As far as I can tell, your linked article defends this on the grounds that this is worth it because it discourages extramarital affairs. But this is a horrible trade-off! The fate of the country is clearly much more important than slightly discouraging extramarital affairs.

At no point is any of the photos obtained illegally. But clearly it is bad for society for the politician to be acting in the interests of the evil party, rather than in the interests of voters.

I mean, the most obvious objection is that there’s no reason why this would tend to help the “evil party” over the “good party”. Are the representatives of the “good party” more prone to engaging in extramarital affairs?

All this seems to do on balance is discourage representatives from both parties from engaging in morally compromising activities.

Second, the consideration here is essentially the same as political bribery: the giving of something of value to a politician in exchange for a political benefit. This may show a difference between something you are not allowed to do for money and something you are allowed to do for free (it is okay to simply ask politicians for favors). But it doesn’t show a difference between blackmail and other forms of bribery, such as the “evil party” simply writing out a $50,000 check to the “good party” politician in exchange for the same benefit.

—

Forget privacy: you’re terrible at targeting anyway - apenwarr (via the-grey-tribe)

“Please point to where I mentioned the repugnant conclusion.”

This is just so obnoxious. If you think there’s a distinction, point it out instead of acting like someone’s an idiot for not noticing it.