Re[7]: 29.03.2023-23:15:57 :: Something smarter than us is going to emerge


node:	Re[7]: 29.03.2023-23:15:57
template:	4
parent:	Re[6]: 29.03.2023-23:15:57
owner:	evad
viewed by:
created:	31.03.2023 - 11:58:39
updated:	31.03.2023 - 12:03:32

cwbe coordinatez:
101
63533
63608
8771344
9052365
9052379
9052380
9052381
9052409
9052480
9052725
9052796

ABSOLUT
KYBERIA

permissions
you:	r,
system:	public
net:	yes

v sirsom horizonte urcite ano
zatial pokial viem AI stale nema intencionalitu (?)

aj ked hned sa da oponovat, ze sice nema, ale ze ju proste nafejkuje, tak ako sa to teraz simuluje v roznych promptoch (imagine you are.. )

dalsi blok je, ze AI nema (aspon dufam) zatial moc pristup k exekutivnym moznostiam, z pohladu svojej existencie (prisun elektriny, schopnost plosne riadit bez ludskeho pricinenia dalsie technologie atp) je stale na nas zavisle a je iba na nas ci a ake bloky vytvorime aby to tak zostalo....

trochu je to teraz podobny pocit, ako kedysi pri zaciatkoch internetu, ze sa deje nieco velke co nas presahuje a zaroven aj nieco co nas moze uplne prevalcovat. vtedy som bola nadsena, ale tentokrat mam fakt uzkost. mam chut si k tomuto precitat / vypocut nieco sucasne (2023), od niekoho kto je zorientovany. nejake tipy?

e: medzitym https://kyberia.sk/id/9052676
presne to som hladala, uz to kukam

00000101000635330006360808771344090523650905237909052380090523810905240909052480090527250905279609053024

uplne rozumiem tym pocitom, podla mna je aj vazny dovod na obavy, ale som toho nazoru ze nie preto, ze by tu vznikala nejaka prilis silna umela inteligencia ktora povedie k nerelevantnosti ludstva.

tu su nejake tipy na zdroje, ak to bola otazka vseobecne a nie iba autorovi prispevku, na ktory reagues.

nie je to 2023, ale je to sucasne, vyborne citanie, a odporucal by som ho uplne vsetkym, vratane vyskumnikom v hociktorej oblasti toho umbrella termu "umela inteligencia" - Melanie Mitchell - Artificial Intelligence: A Guide for Thinking Humans

a nieco kratke z pred par dni - Emily Bender: Policy makers: Please don’t fall for the distractions of #AIhype

Just sayin’: We wrote a whole paper in late 2020 (Stochastic Parrots, published in 2021) pointing out that this head-long rush to ever larger language models without considering risks was a bad thing. But the risks and harms have never been about “too powerful AI”.

Instead: They’re about concentration of power in the hands of people, about reproducing systems of oppression, about damage to the information ecosystem, and about damage to the natural ecosystem (through profligate use of energy resources).

Pripadne vyjadrenie autorov toho vyssie spominaneho stochastic parrots clanku k tomu istemu: Statement from the listed authors of Stochastic Parrots on the “AI pause” letter

no a ten Stochastic Parrots clanok je vynikajuci: https://dl.acm.org/doi/10.1145/3442188.3445922

0000010100063533000636080877134409052365090523790905238009052381090524090905248009052725090527960905302409053136

tieto autorky su na opacnej strane ai-diskurz potkovy ako yudkowsky imho :) woke vs doomer agitka

0000010100063533000636080877134409052365090523790905238009052381090524090905248009052725090527960905302409053038

toto presne som potrebovala, dik moc!!

Stochastic Parrots clanok je fakt zasadny, aj aktualne reakcie jeho autorov.

000001010006353300063608087713440905236509052379090523800905238109052409090524800905272509052796090530240905303809053117

stochastic parrots by sa tak hodil akurat na a2larm, co som zachytil tak z odborneho hladiska je povazovany viac za 'kontroverzny' nez zasadny

00000101000635330006360808771344090523650905237909052380090523810905240909052480090527250905279609053024090530380905311709053122

..dajme tomu, ze ze pre moje momentalne pochopenie sa javi ako zasadny.

mas nieco lepsie?
(bez ironie, dlhsiu dobu som to nesledovala a chcem sa zorientovat, to je cele)

e: btw a2larm financne podporujem uz par rokov :)

0000010100063533000636080877134409052365090523790905238009052381090524090905248009052725090527960905302409053038090531170905312209053135

na vseobecnu orientaciu/ako rozcestnik podla mna gwern a uz nizsie spominany scott alexander, pisu o tom rozumne vyvazene a pomerne zrozumitelne.

00000101000635330006360808771344090523650905237909052380090523810905240909052480090527250905279609052899

tuto je zhrnac "sceny"
https://astralcodexten.substack.com/p/why-i-am-not-as-much-of-a-doomer