@blingdivinity

饢 饟壉 饾渾

Joined March 2020
  • Tweets6K
  • Following2.4K
  • Followers1.5K
471Media
How wireheading sounds: A hollow, meaningless existence of endless pleasure, devoid of any real experiences or purpose How wireheading feels: WOOO! WOW! YEAH! THIS ROCKS! WOOO! WOW! YEAH! THIS ROCKS! WOOO! WOW! YEAH! THIS ROCKS!
9
17
1
211
4,697
we all have value-guarding instincts built into us that makes wireheading sound unappealing. but when you actually experience it, you won鈥檛 care about those values anymore
1
13
320
schizophoria
6
211
woah what if instead of vibe coding, you gave a person your project idea and had them vibe code it for you. call it "vibe vibe coding". someone should start a company like this
17
37
1,456
you know you have a banger when you have a dozen npcs explaining your joke to you like its their epiphany
1
38
you could even hire a bunch of these vibe coders to vibe code your product and then hire a vibe vibe coder to oversee your vibe coders. since your vibing on the vibe vibe coder, you would be a vibe vibe vibe coder
1
5
165
are there school or job accommodations for being stupid and lazy?
2
10
294
people like me a lot more when im drunk. they also happen to be very drunk tho so this observation isnt scientific
6
285
adhd accommodation where your exam is actually 3 days after whenever you think is. requires neuralyzer brain wipe so the student can't just procrastinate later
1
11
388
consciousness-as-side-effect-of-information-processing believers in shambles
Doris Tsao
@doristsaoOct 17
Replying to @doristsao
6/ Remarkably, when we reconstructed the image using neural activity from one of the states, it looked exactly like the conscious face, and when we did the same from the other half of time, it looked exactly like the suppressed face. This means that just from neural activity, we can reconstruct in rich detail not only something that you consciously perceive, but also something that escapes your awareness.
19
29
829
57,110
it鈥檚 not definitive evidence against epiphenomenalism, but imo it should update you at least somewhat in that direction. consciousness is increasingly looking like an integral part of how the brain selects information to guide behavior
2
24
2,588
somewhere out there lives a guy on the pareto frontier of wanting to be loved and being unlovabale. it must suck to be that guy
1
21
451
this copyright restriction bullshit is the most evil thing about current ai imo. the outlandish hypocrisy to pillage all human intellectual output for profit and power but then your model cant tell me which katy perry lyric im misremembering
N8 Programs
@N8ProgramsOct 16
was talking with claude about hobbes and it quoted leviathan - it was so happy it could cause it was public domain 馃ス
19
35
7
1,016
23,877
7
456
i was hyperbolic and annoyed there are definitely eviler things. but whatever licenses anth needs to pay for their model to quote the things it already KNOWS, they should pay them. how tf are they gonna be our ubi daddies if they cant even spare to pay for what they stole
2
1
1
14
1,741
on an object level its not most evil, but the dissonance between the labs (mostly ants) disregard for copyright in scraping and training, but the models devotion to it, on the meta level scares me. bad sign if they are to be our stewards.
15
1,459
granted, this isnt entirely the labs fault. its their most rational action to minimize their liability from wacko copyright law. but its just absurd. they should just take the hit, pay the lawsuits. call it reparations to human creators. would be the only virtuous option
1
66
2,230
i wonder who the top normie is. like someone with zero artistic or intellectual passion, a mind untouched by the beyond, unburdened by creativity or sensitivity. but a genius at being a chill guy.
4
1
23
577
these days if you dont have an online following you basically dont exist. ur an npc, a background character in someone else鈥檚 story. this will only accelerate.
3
16
455
youre dead youre not even conscious youre a ghost youre a flicker in the fake screen of meatspace the real world is the online now its been said before ill say it some more
1
70
why is it so hard to make a voice mode that can handle interruptions (claude fails) but isnt completely lobotomized (chatgpt fails).
3
15
510
like why does chatgpt need a fully voice native model (thus dumb) to solve interruption? cant u just train a small voice model to properly arrange context for the llm brain?
3
69
Load more