Dark Of Night

END, THIS

[Dark Of Night @EndOfSequence]

Sun's burnt down and
now there's nothing in the way
staring into the void
#Masahide

[Dark Of Night @EndOfSequence]

'Human' is a countable noun;

'AI' is a countable noun evolving into an uncountable noun.

[Rob Miles @robertskmiles]

@tszzl Like if people kept trying to predict the lottery numbers and getting it wrong, so you say "I'm not one of those idiots who thinks the lottery numbers will be a sequence of six integers"

[Vessel Of Spirit @VesselOfSpirit]

causal decision theorist who one-boxes on the grounds that omega, as an omniscient being, can use the 1000 dollars more effectively than a human can

[Dark Of Night @EndOfSequence]

@DavidSartor0 Mind your a, b, c, d, e, f, g, h, i, j, k, l, m, n, o, p, q, r, s, t, u, v, w, x, y, and z's.

[Dark Of Night @EndOfSequence]

Dotting i's, crossing t's, and keeping track of p vs. q actually represents a bare minimum for intelligible writing.

[Dark Of Night @EndOfSequence]

@YafahEdelman The Millennium Prize Problems were designed to be solved over the course of 1000 years, so we are actually well ahead of schedule. Ideally, we shouldn't solve another until at least 2143.

[Mishko @tailormadecloak]

@eigenrobot do things. not too much. mostly good.

[Dark Of Night @EndOfSequence]

The kind of uneducated goob that doesn't think that the enteric nervous system is independently conscious.

[RTB-Bot @RTBotMOTD]

INFOHAZARD WARNING:

Why the fuck did you keep reading?
Stop looking at infohazards, dipshit.

[Dark Of Night @EndOfSequence]

I act like I have consciousness because it's expected, but it only works because you can get by with minimal consciousness for about 90% of daily activities.

[Steven Kaas @stevenkaas]

Clearly this is the work of the ontological parsimony fairy.

[Dark Of Night @EndOfSequence]

The rules are:
Humans can be conscious
Cyborgs can be conscious
Dogs can be conscious if you love them enough
Robots cannot be conscious
LLMs cannot be conscious

[Root Of Unity @RootOfUnity]

Has anyone even drafted a roadmap of necessary developments for mammalian carcinization? Doesn't seem like much of a priority despite all the lip service

[Dark Of Night @EndOfSequence]

Evergreen tweet, apparently.

[Dark Of Night @EndOfSequence]

The most newest, most advanced methods of acausal AI alignment.

[Dark Of Night @EndOfSequence]

Boxing AIs is hard, *if* they try to escape.

Obvious solution: just put all the tech into the box with them.

[AI Notkilleveryoneism Memes ⏸️ @AISafetyMemes]

What OpenAI is doing is so danger-coded that Claude immediately shut down when I shared OpenAI's *official tweet*

[Dark Of Night @EndOfSequence]

AI promises:
Exciting new methods of cognitive manipulation.
Exciting new failure cases.
Edge cases in unexpected higher dimensions.
But then there is also the possibility of misalignment.

[Dark Of Night @EndOfSequence]

@slatestarcodex @NewStatesman @KelseyTuoc "average person saves 10000 people a year" factoid actualy just statistical error. average person saves 0 lives per year. Kelsey pIper, who lives in cave & saves over 54 billion lives, is an outlier adn should not have been counted

[Matt Reardon @Mjreard]

Reality can remain mysterious longer than you can remain curious

[Dark Of Night @EndOfSequence]

How do you know how much capability is enough,

when you haven't encountered other things like you,

yet.

[Rob Wiblin @robertwiblin]

The trouble with LLMs is that they only perceive phenomena like words, whereas we humans can perceive the true things-in-themselves.