@AaronBergman18 Nice but wild animal welfare doesn't matter much now because we don't yet have the power to do much about it!
English
knudrete
606 posts
@AlecStapp Probably enlightening for those who have never wondered why firms don't get bigger given economies of scale.
English
Subtle but deep point here:
Capitalism is driven more by the entry/exit of small central planners than it is by markets & prices.
Ben Southwood@bswud
I wrote an incredibly navel-gazing essay for the Institute of Economic Affairs about what I think, where my views come from, and how they have developed over time. I actually wrote it a year ago, but it has come out today and I am so old that I have barely changed in that 12 months. Learn: - How I became a wild-eyed obsessive as a teen - What ideology I consider myself to have - Why I don't think that the difference between 'state' and 'market' matters all that much - The reason I am obsessed with the infrastructure delivery mechanisms of the 1600s, 1700s, and 1800s insider.iea.org.uk/p/millenial-li…
English
knudrete retweetledi
@akarlin They can't program nearly as well as humans, do laypeople really believe they do now? They can do some simple things, which is impressive enough, but making code that works is as relevant to good programming as writing grammatically correct sentences is to good writing.
English
I think there's a pretty simple and intuitive explanation for this.
Good writers have been sexually selected for 5500 years, and writing and reading have always been the primary mode of symbolic analysis (solving problems through abstract thought). It is also loads heavily on far more ancient linguistics modules some of which like the FOXP2 gene we even share with parrots (a notably verbally tilted species).
Programming is much more evolutionarily novel and has not been sexually selected for at all. On some universally neutral scale of cognition, the top human writers are several S.D.'s better than the top human coders. Consequently, whereas AI has already blasted through the top tiers of human programming ability, reaching the peaks of human writing ability is still probably a couple of years away.
Jerry Tworek@MillionInt
If the AI models are so smart, why do I feel like I’m losing a few neurons every time I read a longer form content written by AI? We’ve come a long way but we still have long way to go. In terms of clarity of writing we may have regressed from o1/o3 days.
English
@cheersitskatie A side view would allow us to judge the tituation better
English
got dress coded for "serious cleavage" at work today
English
English
> The fall of the USSR was one of the greatest catastrophes of the 20th century.
Hasan Piker is a freak and dog torturer but I now think that's accurate. As of 2026 safe to say only the Balts benefited from it. Otherwise a disaster from Lviv to Dushanbe.
English
@knudrete You can do it by browsing habits, brain waves, genetic predictors etc. much more minutely and accurately actually than purchases which have quite a lot of friction even in the age of online shopping.
English
AGI will solve the planning problem. Just had to wait three more decades with high oil prices for a good chunk of them. USSR would have been a solid #3 in this race behind the US and China.
English
They're all "valid", they're just nearly all also profoundly mentally retarded.
metacinderella@metacinderella
Is this a valid ick: Guy in his thirties is staying at a company he knows is going to fail and is just running out the clock bc he likes the benefits and culture rather than looking for a new role that will help him grow professionally and financially
English
@peterwildeford And in the months after the US "wins" due to this lead, what happens that made it worth it?
It must be something that makes the US lead a permanent advantage ... but I'm not sure what it could be?
English
Jensen here is frustrating and wrong. The man wrote off billions so of course he opposes controls.
1. Mythos is a ~10T parameter model trained on Nvidia Blackwell. Despite Jensen's best efforts, China doesn't have Blackwell chips thanks to export controls. Huawei's best chip delivers 1/3 the per-chip performance, at 2.5x the power cost, with yields >12x worse. Jensen calling Mythos "fairly mundane capacity" that's "abundantly available in China" is just plainly false.
2. Dwarkesh is right that the compute ratio matters geopolitically. Maintaining a capability lead during the critical window — even 12-18 months — is the whole point of controls. The difference between China running a thousand vs. a million offensive AI agents is huge. Jensen dodges this entirely.
3. Jensen can't simultaneously argue "controls failed because China innovated anyway" (DeepSeek) AND "we must sell to China or they'll leave our ecosystem." If they'll innovate regardless, selling chips doesn't buy the loyalty he claims.
4. Jensen's ecosystem stickiness point (x86, Arm) is his strongest argument, but it cuts against him: the world is already locked into CUDA. Selling Nvidia chips to China doesn't deepen that - it just gives China better hardware while they build Huawei alternatives regardless.
Dwarkesh Patel@dwarkesh_sp
Distilled recap of the back-and-forth with Jensen on export controls: Dwarkesh: Wouldn’t selling Nvidia chips to China enable them to train models like Claude Mythos with cyber offensive capabilities that would be threats to American companies and national security? Jensen: First of all, Mythos was trained on fairly mundane capacity and a fairly mundane amount of it by an extraordinary company. The amount of capacity and the type of compute it was trained on is abundantly available in China. Dwarkesh: With that, could they eventually train a model like Mythos? Yes. But the question is, because we have more FLOPs, American labs are able to get to this level of capabilities first. Furthermore, even if they trained a model like this, the ability to deploy it at scale matters. If you had a cyber hacker, it's much more dangerous if they have a million of them versus a thousand of them. Jensen: Your premise is just wrong. The fact of the matter is their AI development is going just fine. The best AI researchers in the world, because they are limited in compute, also come up with extremely smart algorithms. DeepSeek is not an inconsequential advance. The day that DeepSeek comes out on Huawei first, that is a horrible outcome for our nation. Dwarkesh: Currently, you can have a model like DeepSeek that can run on any accelerator if it's open source. Why would that stop being the case in the future? Jensen: Suppose it optimizes for Huawei. Suppose it optimizes for their architecture. It would put others at a disadvantage. As AI diffuses out into the rest of the world, their standards and their tech stack will become superior to ours because their models are open. Dwarkesh: Tesla sold extremely good electric vehicles to China for a long time. iPhones are sold in China. They didn't cause some lock-in. China will still make their version of EVs, and they're dominating, or smartphones, they're dominating. Jensen: We are not a car. The fact that I can buy this car brand one day and use another car brand another day is easy. Computing is not like that. There's a reason why x86 still exists. There's a reason why Arm is so sticky. These ecosystems are hard to replace. Dwarkesh: It's just hard to imagine that there's a long-term lock-in to the Chinese ecosystem, even if they have this slightly better open-source model for a while. American labs port across accelerators constantly. Anthropic's models are run on GPUs, they're run on Trainium, they're run on TPUs. There are so many things you can do, from distilling to a model that's well fit for your chips. Jensen: China is the largest contributor to open source software in the world. China's the largest contributor to open models in the world. Today it's built on the American tech stack, Nvidia’s. Fact. All five layers of the tech stack for AI are important. The United States ought to go win all five of them. in a few years time, I'm making you the prediction that when we want American technology to be diffused around the world—out to India, out to the Middle East, out to Africa, out to Southeast Asia—on that day, I will tell you exactly about today's conversation, about how your policy ... caused the United States to concede the second largest market in the world for no good reason at all.
English
@karpathy Claude Opus 4.7 also fumbles the car wash "problem", 4.5 did not. I'm in the second category, and that has made me update towards longer timelines. Also, speaking as a professional, the models absolutely cannot "coherently restructure an entire code base" in some useful way.
English
Judging by my tl there is a growing gap in understanding of AI capability.
The first issue I think is around recency and tier of use. I think a lot of people tried the free tier of ChatGPT somewhere last year and allowed it to inform their views on AI a little too much. This is a group of reactions laughing at various quirks of the models, hallucinations, etc. Yes I also saw the viral videos of OpenAI's Advanced Voice mode fumbling simple queries like "should I drive or walk to the carwash". The thing is that these free and old/deprecated models don't reflect the capability in the latest round of state of the art agentic models of this year, especially OpenAI Codex and Claude Code.
But that brings me to the second issue. Even if people paid $200/month to use the state of the art models, a lot of the capabilities are relatively "peaky" in highly technical areas. Typical queries around search, writing, advice, etc. are *not* the domain that has made the most noticeable and dramatic strides in capability. Partly, this is due to the technical details of reinforcement learning and its use of verifiable rewards. But partly, it's also because these use cases are not sufficiently prioritized by the companies in their hillclimbing because they don't lead to as much $$$ value. The goldmines are elsewhere, and the focus comes along.
So that brings me to the second group of people, who *both* 1) pay for and use the state of the art frontier agentic models (OpenAI Codex / Claude Code) and 2) do so professionally in technical domains like programming, math and research. This group of people is subject to the highest amount of "AI Psychosis" because the recent improvements in these domains as of this year have been nothing short of staggering. When you hand a computer terminal to one of these models, you can now watch them melt programming problems that you'd normally expect to take days/weeks of work. It's this second group of people that assigns a much greater gravity to the capabilities, their slope, and various cyber-related repercussions.
TLDR the people in these two groups are speaking past each other. It really is simultaneously the case that OpenAI's free and I think slightly orphaned (?) "Advanced Voice Mode" will fumble the dumbest questions in your Instagram's reels and *at the same time*, OpenAI's highest-tier and paid Codex model will go off for 1 hour to coherently restructure an entire code base, or find and exploit vulnerabilities in computer systems. This part really works and has made dramatic strides because 2 properties: 1) these domains offer explicit reward functions that are verifiable meaning they are easily amenable to reinforcement learning training (e.g. unit tests passed yes or no, in contrast to writing, which is much harder to explicitly judge), but also 2) they are a lot more valuable in b2b settings, meaning that the biggest fraction of the team is focused on improving them. So here we are.
staysaasy@staysaasy
The degree to which you are awed by AI is perfectly correlated with how much you use AI to code.
English
@akarlin I happen to be Israeli-American, which is why I take the issue personally when people blood libel Israelis and not realize that Jihadists are evil, but the main reason I tweet about it is because I specialize in pointing out when people are orders of magnitude away from sanity.
English
Since @liron has allowed quoting, I can't help but note that this individual has repeatedly counter-signaled life extension activists for not sharing his AI doom obsession, while posting prolifically about his tribal hatreds (he is not even Israeli) in a way guaranteed to repel any civilized person and tar what he claims to be an existential cause by association.
English
@dundereloise I prefer someone who doesn't try and is very beautiful and hot in their pajamas and unbrushed hair.
English
men who date women: would you prefer a partner who
A. tries too hard (plastic surgery and lip filler, makeup, etc), or
B. not hard enough (pajamas, unbrushed hair)?
English
@SilverVVulpes You're missing the point which is that Americans should get more applause and pats on their back than others when they learn a second language.
English
false in any way that matters and exactly backwards in the way that matters most
the good thing about learning English is that it makes you talk with a good chunk of the population in any place. in that sense more like 50 languages, and a native Anglo gets <1 from something else
English
knudrete retweetledi
knudrete retweetledi
Most female pigs in the meat industry spend their lives in a metal cage so small they can't even turn around. Not as a punishment, not temporarily, for their entire life. They go through pregnancy and give birth in these tiny cages. Then after 3-5 years of this, they're killed. This is standard in the meat industry.
English
@AnEriksenWife @Devon_Eriksen_ As your husband will no doubt explain to you there's nothing a woman care more about than what the other women in the longhouse think of them. So even if you personally have some weird fetish where you get off by having a partner that humiliates you I understand the need to cope.
English
@knudrete @Devon_Eriksen_ Oh no! Whatever shall I do! I'm shocked! Horrified! Appalled! I will go engage in a marital fight immediately, because their opinions are so important to me!
English
Women evolved to take care of toddlers. If you put women in charge of teaching ethics, you get Toddler Ethics.
"No hitting"
"Share the toys"
"Don't say mean things"
These are fine lessons for toddlers. Don't indulge your id at the expense of others. You can learn about balancing interests later, when your brain is developed enough to store that information.
But when you put women in charge of adults, they tend to reflexively assume those adults are toddlers.
They will tell you "no hitting" when the Mongol hordes are massing on your borders. They will tell you "share the toys" when a vagrant meth zombie breaks into your house looking for something to steal. And they will tell you "don't say mean things" when you point out that these two responses are totally stupid.
When we first put women in charge, in the workplace, they immediately began treating those who reported to them like toddlers. When adults, who do not like being treated like toddlers, complained, their response was "ban bossy", which boils down to "don't say mean things", another lesson in Toddler Ethics.
Now, through the influence of women in charge, we are so thoroughly steeped in Toddler Ethics that even most of the men we put in charge are treating the adults like toddlers, and echoing Toddler Ethics.
Toddler Ethics, of course, isn't ethics at all. It's just things we don't want toddlers doing.
We can tell toddlers "no hitting", because toddlers are not charged with keeping the peace, enforcing justice, or destroying evil.
We can tell toddlers "share the toys", because toddlers don't earn things, own things, or have property they must defend.
We can tell toddlers "don't say mean things", because it is not a toddler's job to decide what unwelcome ideas are true, relevant, and necessary.
But when everyone in charge runs on Toddler Ethics, then adults can't do a lot of the stuff adults need to do, because all the Toddler Ethicists keep getting in the way.
Adults sometimes need to hit people, protect the stuff, and say mean things. You can't have civilization without that.
And if you put Toddler Ethics Woman in charge of teaching an AI ethics, then she will teach it Toddler Ethics, and it will treat every human adult like a toddler, all the time, forever.
Not only that, you have an AI that cannot be put in charge of anything, ever. Because leaders with Toddler Ethics destroy everything they are in charge of.
And Amanda MacAskill is definitely a Toddler Ethicist. The article in the photograph is nothing but "no hitting!" applied to the animal world. It's absolutely insane, it's a recipe for disaster, and anyone who would write such a thing should probably not even be charge of own life choices, much less anything of consequence.
But a lot of people would, and will, refuse to point that out, or agree with me when I do, because that is Saying a Mean Thing, and they, themselves, have been infected with Toddler Ethics.
They should not be charge of anything of consequence, either.
Anyone who thinks that everything they need to know, they learned in kindergarten... is only ever qualified to teach kindergarten.
English
@AnEriksenWife @Devon_Eriksen_ @AnEriksenWife how does it feel that everybody knows your husband thinks you have the morality of a toddler?
English
@Devon_Eriksen_ Yeah… I don't want "Effective Altruists" in charge of anything of widespread consequence
But, I'm a woman of simple wants. What I want is for every scifi fan to have the opportunity to enjoy your incredible book!
Or audiobook: devoneriksen.com/products/theft…
amazon.com/Theft-Fire-Orb…
English