I'm not sure what you think it means. Care to elaborate?

In computer science, provably secure software mechanisms rely on an idealized model of hardware; they aren't and can't be secure against hardware failures.

A crazy prediction: 25 years from now, what we call intermittent fasting will be called a normal daily schedule, and what we call a normal daily schedule (3 meals and some healthier snacks) will be called food addiction. And the primary reason for this change will not be even e.g. obesity but the mental effects: people will consider it an obvious truth that being constantly in a fed state dulls the mind and saps motivation and generates akrasia and generally harms productivity.

They will look back to us and think these people went through life half-asleep because they went through life constantly (nearly) sated.

They will says stuff like "you are a like a dolphin: if you feed yourself before you jumped through all the hoops you planned for that day, you won't jump through them".

Arriving to work with a breakfast in hand will be a bit like arriving to work with a beer in hand: if you roll best that way it is not for others to judge, but most people will prefer to work sober and sharp - and that means literally staying hungry. Today we joke about having a food coma and difficulty to concentrate after a work lunch, fixing ourselves up with coffee: this will sound a lot like as an 1950's person complaining that he finds it hard to concentrate after a two-martini lunch sounds today.

Related SMBC.

The lack of understanding in this comment is depressing.

You say:

"No. The AI does not have good intentions. Its intentions are extremely bad."

If you think this is wrong, take it up with the people whose work I am both quoting and analyzing in this paper, because THAT IS WHAT THEY ARE CLAIMING. I am not the one saying that "the AI is programmed with good intentions", that is their claim.

So I suggest you write a letter to Muehlhauser, Omohundro, Yudkowsky and the various others quoted in the paper, explaining to them that you find their lack of precision depressing.

Short answer is I don't know. The long answer will take a little background.

I haven't bothered to read through Decoy's internals, but this sort of steganography usually hides its secret data in the least significant bits of the decoy image. If that data is encrypted (assuming no headers or footers or obvious block divisions), then it will appear to an attacker like random bytes. Whether or not that's distinguishable from the original image depends on whether the low bits of the original image are observably nonrandom, and that's not something I know offhand -- although most images will be compressed in some fashion and a good compression scheme aims to maximize entropy, so that's something. And if it's mostly random but it does fit a known distribution, then with a little more cleverness it should be possible to write a reversible function that fits the encrypted data into that distribution.

It will definitely be different from the original image on the bit level, if you happen to have a copy of it. That could just mean the image was reencoded at some point, though, which is not unheard of -- though it'd be a little suspicious if only the low bits changed.

Is there good reason to suppose that Gal's desire for internal combustion is irrational

I am trying to steelman this statement, and the best I can come up with is "this particular terminal value of hers is potentially in conflict with some of her other terminal values and could use an adjustment." But I don't know what her other terminal values are.

Do you have plans for when your term life insurance expires, but you're still alive (which is, actuarially speaking, fairly certain)?

That was a great fanfic. The characters were amazingly written, and many of the scenes were genuinely emotional and moving. What's more, the ending actually worked, in a way that I hadn't expected it to work. We didn't get a resolution to the prophecies about apocalypse, the prophecies about ending death, the nature of magic, the nature of time travel, phoenix fire, and many other things... but that actually feels okay. We've been reading an origin story all along, and a great origin story it is. Eliezer, thank you!

That said, now I have a wishlist for some other fanfics I'd like to read:

1) A fanfic where Harry and Hermione use the scientific method to research the nature of magic.

2) A fanfic where smarter versions of canon characters fight each other with complicated plots.

3) A fanfic where most of Harry's successes come from making surprising but rational decisions.

Much of the promise of HPMOR was that it hinted at being all of these things, but now I feel that it doesn't do any of them really well. Maybe my standards have gone up, or maybe the promise is still there to be realized by some other author. Or has it already been done?

...supporters say the opposition leader was assassinated to silence him...

I see headlines like this fairly regularly.

Does anybody know of a list of notable opposition leaders, created when all members of the list were alive? Seems like it could be educational to compare the death rate of the list (a) across countries, and (b) against their respective non-notable demographics.

an atheist turned Catholic

Ditto. :)