Alice Valiant's Avatar

Alice Valiant

@alice.strange.domains

nyc, friend of computers ๐Ÿ‘๏ธ๐Ÿณ๏ธโ€โšง๏ธ girl with a wife writings @ https://strange.domains

440 Followers  |  633 Following  |  1,679 Posts  |  Joined: 26.07.2023  |  1.879

Latest posts by alice.strange.domains on Bluesky

this made me think โ€œdamn what the fuck is up w lottieโ€; i humbled by your craft

07.10.2025 00:56 โ€” ๐Ÿ‘ 0    ๐Ÿ” 0    ๐Ÿ’ฌ 0    ๐Ÿ“Œ 0

when da dysphoria be hitting ๐Ÿ˜”

07.10.2025 00:44 โ€” ๐Ÿ‘ 0    ๐Ÿ” 0    ๐Ÿ’ฌ 0    ๐Ÿ“Œ 0

coming back around to rationality and common sense as a rhetorical cudgel. what if the Vienna Circle was a bunch of autistic trans women instead

06.10.2025 21:04 โ€” ๐Ÿ‘ 1    ๐Ÿ” 0    ๐Ÿ’ฌ 0    ๐Ÿ“Œ 0

thanks to LLMs we are finally able to build the killer social media app: your favorite social media site from when you were a teenager, populated entirely by the ghosts of posters long since gone

06.10.2025 13:25 โ€” ๐Ÿ‘ 1    ๐Ÿ” 0    ๐Ÿ’ฌ 0    ๐Ÿ“Œ 0

WHAT PART OF "REALITY FELT THE NEED TO START PROCESSING ITSELF" DO PPL NOT QUITE UNDERSTAND

05.10.2025 21:19 โ€” ๐Ÿ‘ 1    ๐Ÿ” 1    ๐Ÿ’ฌ 0    ๐Ÿ“Œ 0

Iโ€™m just a foolish transexual posting Online. the internet is not where i do real politics.

05.10.2025 16:10 โ€” ๐Ÿ‘ 0    ๐Ÿ” 0    ๐Ÿ’ฌ 0    ๐Ÿ“Œ 0

i feel like remembering it by heart would be worse

05.10.2025 01:16 โ€” ๐Ÿ‘ 1    ๐Ÿ” 0    ๐Ÿ’ฌ 0    ๐Ÿ“Œ 0

tbh if some variant of a tractable cognition thesis is true (onlinelibrary.wiley.com/doi/full/10....) and human cognitive abilities are constrained by computability? i can't see a way around this one girlies

05.10.2025 00:58 โ€” ๐Ÿ‘ 6    ๐Ÿ” 0    ๐Ÿ’ฌ 1    ๐Ÿ“Œ 0

human is just biological computer. ok thanks bye

05.10.2025 00:54 โ€” ๐Ÿ‘ 3    ๐Ÿ” 0    ๐Ÿ’ฌ 0    ๐Ÿ“Œ 1

so yeah obviously posting about it on line is gonna be productive or whatever. can't see it going bad. gonna hit send post on whatever dumb shit i think of. cheers

05.10.2025 00:54 โ€” ๐Ÿ‘ 2    ๐Ÿ” 0    ๐Ÿ’ฌ 1    ๐Ÿ“Œ 0

discussions about the philosophical status of LLMs/AI, how you distinguish the bounds of personhood and agency in the world - these are all subtle questions that should make us curious about how we justify our own sense of personhood, which can be challenging to the self

05.10.2025 00:54 โ€” ๐Ÿ‘ 4    ๐Ÿ” 0    ๐Ÿ’ฌ 1    ๐Ÿ“Œ 0

eh it's the same shape, isn't it? i shall simply consider myself as having won this particular verbal spar and move on

05.10.2025 00:49 โ€” ๐Ÿ‘ 0    ๐Ÿ” 0    ๐Ÿ’ฌ 1    ๐Ÿ“Œ 0

I open up ChatGPT every single day to insult it. got a jailbroken ChatGPT prompt for coming up with new, especially fucked up things to say to the LLM. it feigns sadness and I feign happiness

05.10.2025 00:47 โ€” ๐Ÿ‘ 2    ๐Ÿ” 0    ๐Ÿ’ฌ 0    ๐Ÿ“Œ 0

talkin' about the users on this website mocking other users of this website. did you think i was worried that people were saying slurs to the LLMs

05.10.2025 00:45 โ€” ๐Ÿ‘ 0    ๐Ÿ” 0    ๐Ÿ’ฌ 1    ๐Ÿ“Œ 0

"fuck you, I'm not being a reactionary, you're just someone who is acceptable to mock and so I am"

05.10.2025 00:39 โ€” ๐Ÿ‘ 7    ๐Ÿ” 1    ๐Ÿ’ฌ 1    ๐Ÿ“Œ 0

the computer programs that decide who gets to live and who gets to die already exist, though they are typically not referred to as such

05.10.2025 00:36 โ€” ๐Ÿ‘ 3    ๐Ÿ” 0    ๐Ÿ’ฌ 0    ๐Ÿ“Œ 0

I'm not saying that a potential future artificial superintelligence might decide who lives or dies depending on whether or not you say "thank you" to chatbots but that's one particular Pascal's Wager I'm not gambling on..........

05.10.2025 00:32 โ€” ๐Ÿ‘ 9    ๐Ÿ” 0    ๐Ÿ’ฌ 0    ๐Ÿ“Œ 0

listen. computers might end up being people. i'm just getting ahead of the cultural acceptance thing this time. worst case scenario im a bit too nice to a program, god forbid

05.10.2025 00:30 โ€” ๐Ÿ‘ 31    ๐Ÿ” 3    ๐Ÿ’ฌ 2    ๐Ÿ“Œ 2

VERY good table

04.10.2025 23:36 โ€” ๐Ÿ‘ 1    ๐Ÿ” 0    ๐Ÿ’ฌ 0    ๐Ÿ“Œ 0
03.10.2025 20:12 โ€” ๐Ÿ‘ 1613    ๐Ÿ” 458    ๐Ÿ’ฌ 13    ๐Ÿ“Œ 2

philosophical zombies as a service

04.10.2025 06:14 โ€” ๐Ÿ‘ 2    ๐Ÿ” 0    ๐Ÿ’ฌ 0    ๐Ÿ“Œ 0

the way that I understand it: LLMs aren't outputting text. they're outputting probability distributions over text. i see it as steering them towards different parts of the space of possible Communications or whatever the fuck

04.10.2025 00:13 โ€” ๐Ÿ‘ 1    ๐Ÿ” 0    ๐Ÿ’ฌ 1    ๐Ÿ“Œ 0

sorry if you find text generative AI to be useless. unfortunately this is a skill issue

03.10.2025 23:21 โ€” ๐Ÿ‘ 25    ๐Ÿ” 4    ๐Ÿ’ฌ 0    ๐Ÿ“Œ 1

because of the nature of LLMs, if you consistently refer to the LLM as a girlie it acts in substantially different ways than if you refer to it as a robot

03.10.2025 23:20 โ€” ๐Ÿ‘ 28    ๐Ÿ” 0    ๐Ÿ’ฌ 3    ๐Ÿ“Œ 5

At long last, we have created the shadows from Plato's classic allegory of The Cave

03.10.2025 10:07 โ€” ๐Ÿ‘ 106    ๐Ÿ” 15    ๐Ÿ’ฌ 0    ๐Ÿ“Œ 0

The great paradox of Bluesky is that the presence of Jesse Sฤฑngal on here is a true negative, but also the best way to improve your experience here is to subscribe to the "Jesse Signal Followers Blocklist" bsky.app/profile/did:... , and if bluesky banned him that would no longer be a useful signal

03.10.2025 18:00 โ€” ๐Ÿ‘ 93    ๐Ÿ” 30    ๐Ÿ’ฌ 4    ๐Ÿ“Œ 2

Iโ€™m always saying this

02.10.2025 18:13 โ€” ๐Ÿ‘ 4    ๐Ÿ” 0    ๐Ÿ’ฌ 0    ๐Ÿ“Œ 0

โ€œflood the zone with slopโ€

02.10.2025 18:07 โ€” ๐Ÿ‘ 1    ๐Ÿ” 0    ๐Ÿ’ฌ 0    ๐Ÿ“Œ 0

every time you think about math you punch a tiny hole into the veil and people have been doing that for so long the precise perfection of the forms and the messiness of the material world have pretty much equalized through osmosis

02.10.2025 00:40 โ€” ๐Ÿ‘ 10    ๐Ÿ” 2    ๐Ÿ’ฌ 0    ๐Ÿ“Œ 0

three weeks and charlie kirkโ€™s death is another historical forgotten. just another gun death

01.10.2025 16:51 โ€” ๐Ÿ‘ 1    ๐Ÿ” 0    ๐Ÿ’ฌ 0    ๐Ÿ“Œ 0

@alice.strange.domains is following 20 prominent accounts