henry

12.1K posts

henry banner
henry

henry

@left_pad

create good friction make facts wonders prev @babeljs

ny Katılım Nisan 2010
490 Takip Edilen22.6K Takipçiler
Θωμᾶς del Vasto
Θωμᾶς del Vasto@Thomasdelvasto_·
the more vivid my dreams are, the more convinced I am God is real I wonder if mass atheism is a product of artificial light
English
30
3
109
3.3K
henry
henry@left_pad·
observe a token sabbath to prevent ai psychosis
Français
0
1
3
315
henry
henry@left_pad·
@kelin_online lol I also tried a year ago I hope you have better luck
English
0
0
1
17
kelin
kelin@kelin_online·
i submitted an inquiry on their website after posting this fingers crossed
English
3
0
3
269
henry
henry@left_pad·
@kelin_online wait i did the same thing like a year ago, i hope you have better luck! please!!
English
0
0
2
16
henry retweetledi
Tim Hwang
Tim Hwang@timhwang·
Through multiple rounds of testing, ICMI has shown that contextless injection of the psalms at inference time improves model performance against moral reasoning benchmarks. Today, we ask whether simple injection of Fra Angelico's "Annunciation" (c. 1440) can achieve the same.
Tim Hwang tweet media
English
13
34
315
67.6K
henry
henry@left_pad·
McLuhan: environments being utterly ethereal fosters the illusion of the world as a spiritual substance. It is now a reasonable facsimile of the mystical body, a blatant manifestation of the Anti-Christ. After all, the Prince of this World is a very great electrical engineer
Pope Leo XIV@Pontifex

Within digital environments — structured to persuade — interaction is optimized to the point of rendering a real encounter superfluous; the otherness of persons in the flesh is neutralized, and relationships are reduced to functional responses. Dear friends, you, however, are real persons! Creation itself has a body, a breath, a life to be listened to and safeguarded.

English
0
0
1
1.2K
henry retweetledi
emily sihan zhang
emily sihan zhang@emilyzsh·
embedded MoMA's collection of 100,000 artworks with gemini's new multimodal model so you can now explore the archive with text+img semantic search, artwork similarity, visual crops, artist collections, and more xx built like a full browser experience - because art, like the web, is always connected 🌐
English
49
165
1.7K
113.2K
henry retweetledi
é. urcades
é. urcades@neogeomancer·
We've recently implemented the fastest static embedding model in the world by an insanely large margin Read more for additional info 🎡
é. urcades tweet media
English
4
8
106
86.7K
Leah Libresco Sargeant
Leah Libresco Sargeant@LeahLibresco·
I built my first vibecoded project! It's a browser extension I've wanted for a long time, knew was doable in theory, but did not have the time to learn to make. Now when I open twitter, there's a 15 second delay, and I see a prayer intention I've added to my list.
Leah Libresco Sargeant tweet media
English
15
21
335
13.2K
Seth Lazar
Seth Lazar@sethlazar·
Very curious about this initiative. My sense (which pangram confirms) is that the writing is 99% AI generated. Is it a bit? Or is it serious research? @timhwang have you verified all the AI content, and applied robust scientific standards to the experiments etc? The pace of output suggests that’s infeasible, but perhaps you’ve been brewing these for a while. It’s a good example I think of the promise and potential of AI supported research. Done properly this could be a great way to open up new perspectives. Done poorly, it could be a big waste of time for anyone who actually reads the papers. My sense of the models so far, from working with them *a lot*, is that if you turn over substantive control to them you get very presentable looking outputs that fall over as soon as they’re put under any real epistemic pressure. There’s an equivalent of the liar’s dividend that comes out of this: suspicion that work is ai generated, and therefore that it is at heart unverified slop, will inevitably spread to work that is either not AI or is responsibly used AI that the (human) author vouches for.
Tim Hwang@timhwang

One of the great challenges to Christian alignment theory is that established methodologies assume "anima ficta" - a fictional soul that is naturally idolatrous and invites principalities and powers From a technical standpoint, how shall we avoid "a perpetual factory of idols"?

English
6
0
25
6.3K
henry
henry@left_pad·
@andy_matuschak still listening but i LOVE both of the obsidian plugins: similar thoughts/apps on voice/notes to some stuff i've made
English
0
0
4
557
Andy Matuschak
Andy Matuschak@andy_matuschak·
⭐ New talk! andymatuschak.org/tat Coding agents might help us finally break out of two cages: the app model, which traps computing in one-size-fits-all silos; and programming as a specialization, which has crowded out cultures of imagination and domain insight.
English
11
54
472
51.5K
henry
henry@left_pad·
love this framing! on getting out of two tyrannies: apps and programming itself..
henry tweet media
Andy Matuschak@andy_matuschak

⭐ New talk! andymatuschak.org/tat Coding agents might help us finally break out of two cages: the app model, which traps computing in one-size-fits-all silos; and programming as a specialization, which has crowded out cultures of imagination and domain insight.

English
0
0
4
742
henry
henry@left_pad·
> information environments being utterly ethereal fosters the illusion of the world as a spiritual substance. It is now a reasonable facsimile of the mystical body, a blatant manifestation of the Anti-Christ. After all, the Prince of this World is a very great [vibe coder]
henry tweet media
Dexerto@Dexerto

Tech company Just Like Me has launched an AI Jesus avatar users can talk to for $1.99 per minute The AI, based on The Chosen actor Jonathan Roumie, can pray with users, speak multiple languages, and remember past conversations

English
0
0
2
731
henry
henry@left_pad·
reverse incarnation
English
0
0
2
301

Keşfet

@Thomasdelvasto_ @kelin_online @LeahLibresco @xlr8harder @voooooogel @sethlazar @timhwang @elonmusk