Claude's performative "this request was sooo dangerous that we downgraded to Opus" is wearing me down I ask it straightfoward questions, nothing shady, nothing complicated, and it tells me it's too dangerous for Fable. What BS.
Claude's performative "this request was sooo dangerous that we downgraded to Opus" is wearing me down I ask it straightfoward questions, nothing shady, nothing complicated, and it tells me it's too dangerous for Fable. What BS.
Evan Prodromou has the unenviable task of making the whole ecosystem easy to understand. Thankfully he does that well. Things are logically laid out, there are comprehensive descriptions of even the most obscure parts of the spec, and it builds up nicely from the fundamentals.
The reality is quite the opposite; a country that can create a national health care system is more aware of its own existence, because of the experience of solidarity which enables everyone to thrive thanks to everyone else.
Then we get to November 24th, 2025. Opus 4.5, to me, was the dividing line, where suddenly... I didn't even try it on the 24th. I think I tried it on the 26th. I give it a couple of tasks, and I realize that the quality of the output is uncannily close to what I would've written.
if you look at the nuclear industry outside of Valor, it's mostly a modeling and simulation uh industry. Like when you think of a nuclear company, right, there's like nuclear companies out there that everyone knows the name of. And you kind of look under the cover. It's actually a modeling and simulation company, right? They produce um very very precise what we call paper reactors which have really good predictions of how a theoretical thing might behave.
Asking an AI for help with the start of an essay, for instance, will invariably shape what direction you end up following, robbing you of the crucial experience of building your own judgement around a topic and selecting your own path to research and argue.
US: restricts US AI model because too dangerous China: builds better Chinese AI model Everyone: switches to Chinese AI model This is why restricting models probably doesn't work, it's much more preferable (for US) to have everyone on US AI models so at least they can monitor what everyone is doing
But if you don't have good LLM intuition, like 100K for smaller models, 200K for these like really beefy like Codeex and Opus 4.8 models is usually a good like training wheel guideline of like if you pass there, your quality of results may be degrading.
Paradoxically, the organization that demands trust can get less of it, while the organization that shows their work - thereby not asking for trust - builds up more of it.
And now it's Everybody's Perfect, a book that pretty much defines what it means for one text to be "in dialog" with another text. In this case, it's Ada Palmer's Inventing the Renaissance, a stunning magnum opus that tells not just the story of the Renaissance, but the story of the story, all the different ways the Renaissance has been used, abused, revised and recovered, starting with the Renaissance itself. It's a book that will make you rethink everything you know about European history, about the world today, and about the very idea of history itself
Ada Palmer may just be the most bewilderingly talented person I know: a genius sf writer, incredible librettist and singer, wildly innovative educator, and a leading historian of the Renaissance, and last year, she published her magnum opus, Inventing the Renaissance, a stunning book about so much more than history
Classroom activities like grammar drills and vocabulary tests feel productive. You are doing something. But none of it builds the network your brain actually needs.
I think the most interesting question is what can the models we have do right now? And so the only thing I care about today is what can Claude Opus 4.6 do that we haven't figured out yet. And I think it will take us 6 months to even start exploring the boundaries of that.
I think Opus 4.5 was the first one that earned my trust. Like I'm very confident now that for classes of problems that I've seen it tackle before, it's not going to do anything stupid.
They could release claw slow mode and have an increase in tokens per dollar by a significant amount. Um they could probably like reduce the price of Opus 46 by you know 4x 5x and reduce the speed by another by maybe just like 2x like the curve on inference throughput versus speed is there already just on hm um and yet they don't um because no one actually wants to use a slow model
But I'm serious, if you're working at a company that uses that gave you Copilot, they think that they're starting to move faster and there's a barbarian horde of people using Opus 4.5 that are going to destroy your company sooner or later.
I initially assumed this was a temporary divide. New users tend to watch closely and check the system's progress, but as trust builds, that scrutiny fades and monitoring starts to feel like a chore. Yet it still seems like there's two camps (for now).
I- i- like, we are in a stage where I'm not building the code base to be perfect for me, but I wanna build a code base that is very easy for an agent to navigate.
Some other comparison is like, Opus is like the coworker that is a little silly sometimes, but it's really funny and you keep him around. And Codex is like the, the weirdo in the corner that you don't wanna talk to, but is reliable and gets shit done.
GPT 5.2 goes till end of August whereas Opus is stuck in mid-March - that’s about 5 months. Which is significant when you wanna use the latest available tools.
This is, bar none, the best sled I have ever used for pushing or pulling. It isn't cheap, but I use it infinitely more than the cheaper sleds I've bought in the past.
I think especially if you have a product that has daily frequency like that's actually the retention that matters the most is that like of your existing user base that has developed a habitual pattern how sticky is your product and it's that retention rate that really compounds and build that builds that daily habit.
You need reproducible builds in order to verify that the app really does what it claims, really encrypts data in a way that it is described on its website. For that you need to make your apps open source for any researchers to have a look at it.
So the tank generals received the Halte Befehl, the stopping order. They didn't believe it when they received it because this would have been a complete victory over Great Britain. This would have been the end of Great Britain. The whole British military was encircled, but they did get out through Dunkirk.
I exclusively use the cheaper Sonnet model. It’s perfectly adequate for my needs, and in fact, I prefer its outputs over the more expensive Opus model.
There's this very good quote from Sam Altman who... He can be a hyperbeast sometimes, but one of the things he said, and I think I agree, is that superhuman persuasion will happen before superhuman intelligence, right? And if that's the case, then these things before we get this AGI ASI stuff, we can embed superhuman persuasion towards our ideal or whatever the ideal of the model maker is, right?
However, productive collaborations are not inevitable. Just look at ethnic violence in South-East Asia or Hindu Nationalism in India. Nation-building is really valuable, learning from multi-ethnic, democratic Indonesia.
And then you can go to chemical system. I don’t think that’s good either. I think there’s a confusion because life emerges in chemistry that life is chemical. I don’t think life is chemical. I think life emerges in chemistry because chemistry is the first thing the universe builds where it cannot exhaust all the possibilities, because the combinatorial space of chemistry is too large.
China built half of the world’s ships (by gross tonnage) in 2022, while the US had 0.2 percent of capacity: in practice, this meant that while China builds hundreds of new ships a year, the US builds three to five.
But they're fundamentally like you're sitting, you're still. I mean, people are just not meant to be that way. I mean, I think you and I have this shared passion for sports and martial arts and doing stuff like that. We're just moving around. It's so much of what makes us people is like, you move around. You're not just like a brain and a tank. It's where the human experience is a physical one.
You can’t change the United States from a country that builds subways for $2 billion/km in New York and $1 billion/km elsewhere to a country that does so for $200 million/km if all you ever do is talk to other Americans.
There is essentially zero correlation - the metros with the largest rent increases had added population / added housing ratios no different than metros with smaller rent increases.
A korrent is a belief a person has stated in their own words: one
sentence stating the claim, backed by a quote and a source, kept at
korrents.com.
Under a name here, the quoted block is what they actually said.
The korrent beneath it is the claim those words support, in
korrents' wording — tap it to see the record, its source, and who
else holds it.
Nobody here wrote their own korrents. They are compiled from public
statements, and a person can change their mind, which is recorded too.
About the English under a post
Some people here publish in a language other than English. Where they
do, this site shows a machine translation beneath the post, in
this typeface — the site's own, not theirs.
The post itself is never changed, moved or hidden: what is set in the
serif above is exactly what the person published, and it is what to
quote them on. A translation can be wrong in ways that matter,
especially about tone.
Only the post's own words are translated. A quoted post, a linked
article and a belief on korrents.com
are left in their original language.