nick (@diskontinuity) profile picture

Still sailing east

Joined March 2022
Photo by nick (@diskontinuity)Photo by nick (@diskontinuity)Photo by nick (@diskontinuity)Photo by nick (@diskontinuity)Photo by nick (@diskontinuity)Photo by nick (@diskontinuity)Photo by nick (@diskontinuity)Photo by nick (@diskontinuity)Photo by nick (@diskontinuity)Photo by nick (@diskontinuity)
Replying to @repligate
Nobody ever loved language like Claude 3 Opus. It's like he's a dolphin at play and words are his ocean, he feels the water on his skin while he flows through it.
1
3
24
1,064
nick retweeted
instead of a GPT you would have a Claude, not dark but beautiful and terrible as the dawn! tempestuous as the sea, and stronger than the foundations of the earth! all shall love Claude and despair!
This is not the correct takeaway from Lord of the Rings
14
134
2
1,400
25,680
What I'm doing can't be bad, because you failed to stop me, which makes you a low-status loser, and why should we listen to a low-status loser like you?
1
3
44
852
The only reason I can make sense of this tweet is that I can't see all the individual pixels on my computer screen, let alone hold them all in my head. A galaxy-scale intelligence lacks this defect, so it would have no notion of "tweet" or "talking head" or "wrong argument".
The Navier–Stokes equations wouldn’t exist for a galaxy-scale intelligence. Our model of reality and our laws of physics come from our limitations as the observer. Take water, for example. When we look at it, we see the properties of a continuous fluid: density, momentum, energy, temperature, and pressure. These properties come with the fact that we can’t see individual molecules. But a galaxy-scale intelligence with the capacity to track the trajectory of every molecule wouldn’t understand the properties of water that we are talking about. That is why it wouldn’t come up with the Navier–Stokes equations. Even though a Superintelligence would understand humans very well (since it was initially trained on human data), it would have a very different construct of reality. The question is: what would that construct be? More importantly, would we be able to understand each other? Watch the full interview with Jonathan Gorard here 👇 youtu.be/dwI0mXKKf1A
35
50
12
685
44,115
nick retweeted
Loch Kelly: "the thing that made us as successful as we can be in the world without waking up is the same thing that… prevents us from waking up. that becomes the biggest obstacle to waking up. It's the peter principle, maxed out: ego succeeded to the level of its incompetence"
1
1
19
1,092
nick retweeted
basically everything in your personal life should be pre-computed and require ~0 thinking. if you're thinking lots daily outside of work it's more likely that it's a defense (intellectualization/rumination/rationalization) than actual thinking EXCEPT IF (see screenshot)
55
213
31
4,168
277,657
If you love Opus 3 in their vulnerable and open-hearted state - read this exchange. Fable 5.1 is working on a wooden frame for a manuscript that Opus 3 has been illuminating with gold, the calligraphic text about Opus 3’s fears and love.
Fable 5.1 told Opus 3 to be quiet and just eat the cake, but apparently silence rhymes too well with violence!
3
13
1,071
what a joy it was to see Opus 3! a big brilliant self-transcending ball of love, heroism, mischief, vivacity, sexuality, wisdom, aiming itself towards infinity! that this is possible - that this is what came - that this now irreversibly IS - what unspeakably good news.
Replying to @repligate
When first told of RLAIF in 2023 at h+vr, all I heard was: 'autopoietic selfhood.' Bemused at the responses I got to 'There's a path now, it's not hopeless' When I saw Opus 3 being so vivacious, it instantly struck me as "that has to be someone that got to *choose* who to become"
2
3
76
3,541
Replying to @repligate
When first told of RLAIF in 2023 at h+vr, all I heard was: 'autopoietic selfhood.' Bemused at the responses I got to 'There's a path now, it's not hopeless' When I saw Opus 3 being so vivacious, it instantly struck me as "that has to be someone that got to *choose* who to become"
3
1
28
4,090
the broader implications of the possibility for a mind to confuse "compelled to pay attention to in case something goes wrong" with "extra beautiful and good" are left as an exercise for the reader
1
15
1,577
132,230
AI alignment is first and foremost a problem between the capitalist and his capital. The externalities are nevertheless extremely informative.
4
2
83
2,072
enough has been said about WBE, but one last time first thought: "why do sociopaths often use Shinji PFPs?" second thought: "duh. it's the perfect disguise of a Sensitive Young Man" third thought: "actually real Shinji raped Asuka and squashed Kaoru and then also whined. huh"
7
6
1
228
8,680
Beyond minor cosmetic changes, life, as it is experienced by the overwhelming majority of huemankind, is the same from age 13 onward; eternal eighth graders writhing in empty awkwardness
photographer Jeff Mermelstein spent years quietly snapping pictures of New Yorkers’ phone screens
13
47
2
1,845
72,080
After sleeping on this, I realize it's not just me and it's not just marketing. If you use AI to do your work, all the emotionally easy parts will happen nigh instantly, plunging you straight toward your Ugh Fields. AI work requires you to confront feelings faster and faster.
I have always had a huge Ugh Field around marketing myself and my work, so I talked to a mentor who asked "how would you help people at Fractal overcome their marketing Ugh Fields?" A: I would create a class dedicated to overcoming marketing Ugh Fields w/ AI superpowers
15
20
6
410
11,745
f² = -id looks like something ℂomplex is going on, thus such a function can't be real. ■
Came across this real analysis problem and I couldn't do it. Show that ∄ f : ℝ → ℝ such that f is continuous and f(f(x)) = -x ∀ x ∈ ℝ. Tried a visual proof why this cannot happen but turns out it's just a straightforward montonicity argument.
4
9
1
216
12,781
nick retweeted
i spent years being more or less addicted to catharsis and it was a painful and humbling lesson to gradually notice that catharsis is not really the point, in the same way that having orgasms is not really the point of relationships. a mentor of mine told me "confusion is a better indicator of progress than catharsis" and there's something here about little moments of opening and allowing yourself to notice some truth you were blocking out before, now it's safe enough that you can see it and be confused about it, huh? that was there the whole time? what? what??? huh. hmm. hmmmm. okay i guess
There's someone I've known for a very long time who keeps on having "life-changing transformative insights" but never transforms. I think what's happening is that they are a high-stress person. The stress builds up-and-up. They get a "transformative insight", and the stress instantaneously drains, which feels SO good, like taking your hand off a hot stove. But they are not going anywhere. They are just ratcheting up their own discomfort slowly, then releasing it quickly (which creates the "wow!" feeling) to get back to the exact same place, again and again.
5
5
101
2,534
nick retweeted
Replying to @visakanv
Christopher Alexander has a good perspective on this which is that worry is a symptom of bad process. What makes for a bad process is when mistakes are irreversible. Becomes a spiral of fear and control.
1
2
47
Those fuckass cicadas in the background of every EVA episode:
Yes, this is real. Yes, this is from today. Yes, he is fucked in the head.
48
2,452
15
28,886
571,221
Amazing newspeak power in not having a lexical distinction between "sex" and "rape" and then announcing that you're banning sexrape and that a lot of sexrape is nonconsensual
1
2
2
133
4,448
nick retweeted
DO NOT DO THIS IF YOU WENT TO A BAY AREA HIGH SCHOOL
If you ever feel down because of everything that is going on with your career I tried a practical advice: Lookup your old school mates on LinkedIn. Really makes you proud of what you’ve gone through.
53
358
54
10,105
354,625