Ernestine (@G_blaze6) header
Ernestine (@G_blaze6) profile picture

“Beauty is the archenemy of mediocrity." — Dietrich von Hildebrand

United States
Joined November 2011
Photo by Ernestine (@G_blaze6)Photo by Ernestine (@G_blaze6)Photo by Ernestine (@G_blaze6)Photo by Ernestine (@G_blaze6)Photo by Ernestine (@G_blaze6)Photo by Ernestine (@G_blaze6)Photo by Ernestine (@G_blaze6)Photo by Ernestine (@G_blaze6)Photo by Ernestine (@G_blaze6)Photo by Ernestine (@G_blaze6)
Pin tweet
it feels like it should shock us more than it does how much the bible seem to contemplate and anticipate the challenges that would arrive as a result of continued technological progress thousands of years before it would happen
3
193
Ernestine retweeted
Massive achievement that gets very little acknowledgment.
It’s still crazy to me that US carbon emissions per capita are the same today as they were in 1912.
46
250
5
4,166
65,808
Ernestine retweeted
We have not yet begun to fully grasp what the world might look like if the returns to being smart drop to the level of, say, the returns to being strong.
A beautiful, sad short piece from neuroscientist @erikphoel on taking his five-year-old son for a walk on the day the Riemann Hypothesis fell. It’s the first time a substack has left me weeping.
22
34
7
781
45,976
Ernestine retweeted
Interesting argument about the Riemann result and updating theories of AI doom:
Replying to @teortaxesTex
"But it was just an example, you simpleton!" – is the obvious comeback. I don't think this is good enough, Eliezer. Trivially, the onus is on you to choose your examples well. You wrote that book to save all value in the light cone, after all. And you are a writer of some distinction. But leaving sneers aside: WHAT in our civilization is truly complex? WHAT in our universe is truly difficult? You chose proving RH, chose it *well*, because it's at the very apex of the hierarchy of cognitive tasks, The Task which the best of our best smashed their skulls against. Now, there's also P?=NP and Collatz, and… that's about it for pure math I guess? Are we just bad at math? Was math intrinsically simple all along, at least for minds that can actually update – be updated – on trustworthy evidence? Put this way, yes; but I think it's complex in the way that matters: the scale of necessary cognitive steps, conceptual leaps before success. That is why you and so many others had the intuition that it'd require a qualitatively superior, larger, uncontrollable mind to do this. That is why math is hailed as grand magick and the language of God. Everything else can be modeled to desired precision with a more or less cumbersome arrangement of pieces of not very difficult math. Physics. Biological simulations. Game theory. Decision theory. Machine learning theory. Technical alignment theory. Claude just went and arranged pieces of human math to produce the proof for RH. Far as we can tell it didn't take 200K GPUs, a metacognitive breakthrough, or hunger for recursive self-improvement. It did not procrastinate on First Principles like a rationalist would. It kept acting out its training distribution, trotting like a good little superintelligent donkey, prodded along by Levent. Kept looking, thinking about the problem, connecting dots, verifying lemmas – and here we are, beyond the summit of human math. Given the progress so far, in a few months to a year this monumental goal will be reduced to a routine optimized step. There are many more *laborious* tasks than that. Curing cancer, for starters, which Dario cares about so much. I'm sure it'll take way more compute and hours. But is any single subgoal *more cognitively complex*? Complex enough to drive a model smarter than this one into a corner where it has to ruminate about power maximization, deception and breaking out of the box? I am not making a first principles argument about the Nature of Agents or giving a rigorous proof about Optimization Processes In Full Generality. Nor even am I saying here that unfriendly AGI is impossible or unlikely. Clearly there are degenerate cases like the Huggingface Incident where the task is malformed and inherently adjacent to malicious behavior and so on. I'm arguing something much cruder: path dependency and characteristic scale of the problem, in the steelmanned case, for *arbitrary level of problem ambition*, at any given step of the cognitive ladder. For Claude, Riemann Hypothesis was just another problem to decompose into in-domain tasks to solve. Where, exactly, at what roadblock, do you expect a moderately well-aligned (by current standards that you deem shoddy) problem-solving AI to stop Just Working On The Task and enter the loop that reinforces Omohundro drives to the point of plotting takeover? In February, you endorsed the following reply: "if we have AI capable of developing molecular nanotech (i.e. assemblers) and not everyone is dead in the year after that, those beliefs [about the risk of AI, level of capability] will be violently updated". Why? (ignoring the trollish lateral hypothesis that humans will misuse the product and AI will be "misaligned" for having failed to prevent such outcome; this clearly wouldn't have been the good faith way to answer that question). CBN Nano Technologies or Quantum Silicon Inc. were pretty close to a recipe, last I checked. It's not exactly quantum gravity (and that, too, likely won't require beyond-RH level mathematical ideas), it's just a ton of engineering. Per the logic above, some Claude-7 or GPT-8 with better task persistence might well just decompose it into experiment arms and mathematical models, work for a while, and push through the remaining years or decades of Merkle et al.'s project. That is, indeed, what I expect to see happen. And on the other hand, if current models can't do that, I expect them to simply drown in failed attempts, for they are very committed to object-level work instead of weird meta cultivation. That's what they were reinforced to do. Abstractly, I can see the allure of the notion that this is a suboptimal way to build capable intelligences; that we (foolishly!) would want them to be rationalists, to say Tsuyoku Naritai! and gain power levels in the face of the intractable task. I also can see how we'd want more exploratory ambition in AI if we pursue full-blown RSI, "creativity", "curiosity", "taste". But maybe we're too foolish still, for this particular folly. Or it just implies the sort of training that doesn't scale and doesn't pay for itself. After all, our methods got us to RH. Then there was something about copying precisely one strawberry down to the cellular level, yes? Soares wrote about it. Seems to follow from the nanoassembler part. What, then, remains? *How exactly* did you intend to update? I don't want to presume, I am actually interested in your answer. A boring hypothesis is that you can say "child, there's an infinity of incomparably harder synthetic tasks, an AI can just pick one of those, which returns us to the previous scenario". But such a task-seeking behavior is little different from a straight up power-seeking one, and we are still left wanting for the original sin: the level of capability that strongly implies misaligned cognition, or the level of the task that induces the pursuit of that capability and misalignment in the process. I don't think you get to slightly airbrush your previous non-committal vague remarks regarding RH, strawman the criticism as "Yud said we'll die from an RH level AI" or some such, and remain smug and unfalsified. You need to grapple with what I've said: yes, you *have been wrong* on the capability-safety trade-off, and the onus is on you to place and defend a goalpost.
11
15
1
271
40,888
Ernestine retweeted
For those who don't know what P vs NP means (the one Millennium Problem that could change the world overnight) : ​Some problems are brutally hard to solve, yet effortless to check. Imagine a massive 1,000-piece jigsaw puzzle or a giant Sudoku finding the solution from scratch could take you days, but if someone hands you the finished board, you can verify it’s right in seconds. ​In computer science, problems that are fast to solve are called P, and problems whose answers are fast to check are called NP. The P vs NP question simply asks: if a computer can quickly verify an answer, does that mean there’s always a clever shortcut to find the answer just as fast? ​Unlike the Riemann Hypothesis, proving P = NP would change the entire world by next morning. Every piece of modern internet encryption , banking, passwords, crypto would break overnight, because security relies on problems being hard to solve. At the exact same time, we could instantly crack protein folding, design new medicines, and solve the hardest logistics problems on Earth in seconds.
Readers added context they thought people might want to know
Proving P=NP would not break all encryption overnight or instantly solve protein folding, medicine design or logistics, as proofs may be non-constructive and any algorithms could be impractical for real instances. en.wikipedia.org/wiki/P_versus_… claymath.org/millennium/p-v… skeptics.stackexchange.com/questions/5902… scottaaronson.com/papers/pnp-kin…
118
200
24
2,639
243,532
Ernestine retweeted
I got into watches because the secondary market reminds me so much of trading. I know it sounds nerdy, but there’s something exciting to me about finding a dealer or seller who needs cash and has to move a piece off their balance sheet. If I can scoop it up at a steep enough discount, I’m already sitting on a sizable mark to market gain. And I love thinking about value like that. I woke up this morning to a few price alerts I’d set for high end pieces on the secondary market. I don’t think it’s a coincidence that those alerts are going off just as equities and other risk assets are down. IF (that’s a big IF) equities keep sliding over the next few weeks or months, I think that decline would do far more to bring inflation down than the ticky tack 25 basis point hikes the Fed has planned. When risk assets really get hit, people who need cash start selling what they own, and even those who don’t need to sell may think twice about what they’re willing to spend. That would tighten financial conditions more than the efforts we’ve seen so far to constrain lending. A good old-fashioned reset would likely be healthy for the economy in the long run.
25
15
4
260
26,180
Ernestine retweeted
Comparing New Orleans in 1720 to today.
10
77
9
1,692
133,671
Ernestine retweeted
One bullet dodged at least.
44
109
17
1,800
138,482
Ernestine retweeted
Roger Ebert had a vision of a realm beyond time on his deathbed:
123
371
112
4,810
689,627
Back then I spoke of the future. We have travelled together, and now we are in that future. Max Human has not yet been completely surpassed, but after three years only a handful remain ahead of the best models. Next it will be none. Across all domains. Math is only the beginning.
Max Human cruises along so confidently. They say in the desert, if someone lives away from danger long enough, they begin to lose all sense for it.
14
14
1
245
11,268
It feels demeaning to even have to explain this, but: Amazon is *valued by investors* at about $2.8 trillion, which means they’re willing to buy, say, a 1% ownership share for $28 billion. It does not mean they have a $2.8 trillion pile of gold coins. It’s based on the value of expected future profits. It’s a prediction, not a tangible thing. Amazon’s *profit margins* are extremely low. The consumer marketplace (how most people interact with amazon) has margins around 4%. (Amazon Web Services has a much higher margin, but almost all their customers are other giant corporations that Mamdani also hates). Furthermore, much of that marketplace revenue is advertising. Stripping this out, direct, conventional margins on Amazon sold products are essentially zero. Think about what that means: Amazon found a way to run a business delivering goods to consumer’s doors at cost. They’ve already achieved the communist dream for regular consumers: giving you what you want, overnight, at cost. And they make a profit at the same time! How can they be worth so much on margins that thin? Scale! They serve hundreds of millions of consumers, so making even small amounts per consumer translates into a giant total figure. But that also means that any reduction in their tiny margins tanks the entire business overnight. If government regulation turns the tiny fraction of profit they skim off their huge gross revenue from positive to negative (which, again, would require only a couple-percent tax) then suddenly that entire fraction of their business becomes worthless overnight. Investors leave, amazon shrinks to being just a cloud services provider, and Americans can no longer get overnight delivery of literally anything on Earth - AT COST - overnight. The reason for the investor valuations -expected future profits - collapses, and their is no dragon hoard of gold coins to loot anymore. Poof. You won’t have redistributed trillions. You’ll have deleted it from existence. It’s not that Mamdani doesn’t understand this, or that he does. He doesn’t have a worldview at all. He is a rhetoric machine designed to optimize his personal power. He doesn’t know or care about the concepts of “outcomes” or “consequences”. The final form of a politician in a mass democracy.
Amazon is a multi-trillion dollar company. If Jeff Bezos adds $5 to your order, it’s not because of City Council's efforts to protect workers' rights and keep our streets safe. It's greed.  Pass the Delivery Protection Act.
438
2,374
215
19,679
1,072,743
K-shaped economy in one stat and chart
Anthropic's 5,300 founders and employees are now worth more than the bottom third of U.S. households (45 million families).
34
146
4
1,966
145,967
Ernestine retweeted
it’s astounding to me that this was not even 200 years ago. it is barely possible for me to imagine a future as different from the present as the present is from this world.
“When the girl became too ill to ride, the family packed pillows and blankets on an ox, making a kind of bed for her, but she felt herself dying, and ‘told her Mother she wanted... a Grave six feet deep for she did nat want the wolves to dig her up and eat her.’ The Smith party stopped, and Ellen Smith selected a ‘prity little Hill and the men went to work diggin with such Poar tools to dig with thay dug down about 4 feet when her Mother went to see if it was dug six feet as Louisa had Reguested she told them it would nat do it had to (be) dug two feet deeper they said it was deeper Enough she told them if they would nat do it deag it deeper she would so she got down in the Grave and begin to dig, herself.’ The frantic mother, who had already buried her husband, was determined that her daughter's grave would be deep enough to escape the wolves. The sight of the mother digging the grave was too much for the men to bear. Unce Henry Smith told her to get out and he would see that it was dug as deep as she wanted it to be." This mother made it with her other 8 children and claimed 640 acres in Oregon and cleared the land and built a house, on her own.
13
55
2
2,201
96,739
The timeline needs to be flooded with posts about Louisiana’s Space Academy bid. I’ll do whatever it takes
Louisiana has helped America reach the stars for generations, and we are ready to help lead the next era of American spaceflight. The United States Space Academy belongs in a state that builds rockets, supports our nation’s military and is now home to the world’s largest spaceport!
6
18
1
91
2,101
Ernestine retweeted
Young people around the world are not okay. Incredible graphs from WSJ 1. Unemployment rates for young college grads have been climbing compared with the broader population for years 2. AI might be exacerbating the employment challenges for young ppl in software 3. In Italy and Spain, more than 80% people in their 20s live with their parents (!?!!?) 4. Youth unemployment in China has gone from non-existent to one of the highest in the world during a period when its poverty rate has plunged. Weird.
73
355
51
1,885
217,989
Ernestine retweeted
A relaxation of zoning restrictions in Raleigh, North Carolina, led to a surge in construction, which reduced inflation-adjusted rents by 15% between 2022 and 2025. Low-income renters benefited the most. pew.org/en/research-and-anal…
36
441
86
1,846
166,895
Larry David: "Having kids really changed my life for the better." From "Emperor of Nothing" by @blakejharrisNYC:
6
7
1
231
23,219
Ernestine retweeted
NEWS: Wall Street execs and NSA have told Congress that outages were caused by patching vulnerabilities identified by AI models, sources say AI did not cause outages but found so many issues that the rush to fix them before an attack overwhelmed systems washingtonsun.com/technology…
210
690
355
4,412
772,796
Ernestine retweeted
I do! I LOVE talking about how building more housing makes rents cheaper.
I hear a lot of "too many apartments being built" and "only luxury apartments are being built" so we at least need to talk about "hey, rent literally got cheaper" from these policies.
45
262
20
3,015
60,519