I asked someone after the 9/11 attacks about the possibility of the USA invading Iran and even back then I got a "lol no that's nuts the USA would have its arse handed to it" kind of answer. Better phrased, but basically that.

There are fallback mechanisms when the risk is per model provider (as in, "What if the price shoots up 10x or Claude goes down for a day" is a manageable concern), but I'd be more worried about the way all models regardless of provider have similar failure modes, i.e. that some tasks fail in similar ways for all models. In some ways, LLMs are collectively like Star Trek's Borg: you've met one, you've met all of them.

The current complacency, one which we are currently still in the process of unwinding from (it will take years) is that of trade turning violent enemies into mutually beneficial growth opportunities. Russia was the first wake-up call there (but even then for the current situation not for Crimea), and over the last year also the USA. China is, I think, currently mostly seen as opportunity rather than threat.

War is expensive, and not doing it is good when possible. It is bad for everyone that we now feel the need to put 5% or whatever of our GDP into defence when it could have been spent on infrastructure, education, healthcare, or even startup grants.

While the US is capable of levelling all settlements, let alone cities, in Iran, it would be an extremely Pyrrhic victory. Like, oil would rise to $200 as a baseline, with occasional spikes at $300, US general inflation would gain 3-7% over baseline (food in particular 25% or so), and piss off all other trading partners worldwide, which amongst other things will make European nations transition even faster to renewables and nuclear using stuff they buy from China and make locally rather than from the US because they actually export useful hardware while the US mostly exports end user licence agreements and what little hardware it exports is itself heavily dependent on China and we can cut out the destabilising middle-man.

Given how many European nations rejected US requests for base/airspace use even with this conflict, a total war against Iran would probably have the US asked to vacate all existing bases in Europe. Even if the US doesn't leave NATO it will become a redundant organisation due to all other members making a new club without inviting the US.

And that's even if the US military obey illegal orders rather than their oaths, given the end of the Iranian civilisation would necessarily involve war crimes.

Still, more of my stuff is moving away from the IDE and to the command line anyway, given agentic coding. I do wonder how long even that will allow me to go without being forced to downgrade to Liquid Glass?

I read it like I always read the GPT-2 announcement no matter what others say: It's *not* being called "too dangerous to ever release", but rather "we need to be mindful, knowing perfectly well that other AI companies can replicate this imminently".

The important corps (so presumably including the Linux Foundation, bigger banks and power stations, and quite possibly excluding x.com) will get access now, and some other LLM which is just as capable will give it to everyone in 3 months time at which point there's no benefit to Anthropic keeping it off-limits.

Not that I wish to anthropomorphise it in this answer, but businesses have managed just fine when humans do this for "lunch breaks" and "going home for the evening to sleep".

(And even mandatory meetings which should have been emails).

I'm of the opinion that while e.g. xAI is in a pump game, OpenAI is at least trying to make money. But even if they're not, even if the DCs are as you say "a financial/political vehicle to pump the markets", they can still be physically real things.

That said, I have no idea how close to complete the Stargate UAE site is.

Shame, in a democratic system, is good for exactly the same reason it is good for individuals to be able to feel it.

Every system of government has a failure mode. There are no exceptions: anarchy fails to strong men with big sticks, democracy fails to demagogues, hereditary aristocracy to literal inbreeding, dictatorships to sycophantic courtiers, etc.

A democracy that one cannot be ashamed of is simply one in which the same demagogues can keep trying as many times as it takes.

I know what you mean. I guess we'll all find out in a few hours what the payloads are, given that while B-52s are nuclear-capable, that's not the only thing they're used for.

And I guess also, when it is a pick-one dichotomy, whether the USAF obeys either their orders or their oaths.

To an American, race may be e.g.: {White, Black, Hispanic, Native American, Arabic, Asian}.

To most Europeans, everyone who an an American would call Hispanic, we'd probably either call "Caucasian" (i.e. white) if their heritage is more from the Spanish side or Native American if their heritage is more from the pre-Columbian (Aztec?) side.

If you're Chinese, they may say the ethnicities are "Han, Zhuang, Hui, Manchu, Uyghur, …" where those are all ethnic groups within China.

Rwanda, infamously, would get you a distinction between Hutu and Tutsi; if you show me a picture of two people and ask me which was which, I wouldn't be able to answer, or even know if I was being pranked with any of the other ethnic groups in Africa.

But more broadly, while skin colour is easy to spot from the outside, it's about as useful as hair or eye colour when it comes to correspondence with the huge range of invisible genetic variation.

I'm not sure which of 1 and 2 is least-bad. All depends on downstream consequences, because we're already past the point where everyone's looking at Trump (not just in Iran but also, and we already had this to an extent with Putin attacking Ukraine) and thinking they need a credible deterrent. OTOH, the USA getting suckered into a drawn-out war with Iran in the same way Russia is with Ukraine may be good for almost everyone else, because an exhausted USA is a manageable threat, in a way that the current USA almost certainly isn't.

I currently believe this was accidental on his part, that he was manoeuvred into overpaying by the previous owners.

Nevertheless, I would still count Musk as "intelligent". Not as intelligent as his boosters like to claim (obviously, given the boosters treat him as a god amongst men), but intelligent.

> You'd think standing on a dais and thinking you're Hitler reincarnate would have been enough to make tax-payers rise up; clearly not.

Well, they literally did: https://en.wikipedia.org/wiki/2025_Tesla_vandalism

While genes must play a part in this (if they didn't, all non-humans would also share our IQ*), genetics shift on a much slower timescale than the entire history of IQ tests.

* This pattern matches to the Motte-and-Bailey rhetorical technique, ergo I am suspicious of people who try to tie genetics and IQ until they're clear they're not making a racially charged claim. Last I checked, there is no real evidence that human races are a meaningful genetic category, let alone that anything usually described as "race" correlates to any genes connected to IQ scores.

Isn't one of the bigger problems with ChatGPT that it's much too supportive of whatever the human is talking about?

One thing that's an open pair of questions for me is: exactly how dangerous is a high-altitude EMP anyway? And do countries with nuclear weapons actually model this in war games?

I've seen it suggested that even a relatively modest HAEMP would be able to physically damage most transformers in the continental USA, necessitating replacement of all of them at the same time when they're all custom and have month-to-year lead times. No electricity, no fuel pumps, no refrigeration, in the US 90% of the population starves within a year.

The US power grid could have been defended in the decades since the March 1989 geomagnetic storm revealed such weakness, but given various evidence such as the fires in California caused by failure of century-old power lines that were well past due for repair, I doubt they have been.

The EU have very deliberately avoided actually confiscating Russia's money. It is sill there, and (much to the annoyance of those who do want to actually confiscate it) Russia can have it back as soon as they stop their illegal invasion of Ukraine.

Closest the EU got to actually confiscating that money is, in fact, a loan such that if the peace agreement fails to include reparations then EU nations cover the loan but everyone's hoping Russia agrees to reparations so they don't need to.

Absolutely agreed on all of this. There's problems which you can solve by organising a bunch of juniors, and there's problems they can't do. Even if my experience is typical, which I accept it may not be, Claude is still not a senior developer. That said, I wouldn't phrase it as "with deep system context", the AI are superhuman at context, they're just kinda… weirdly off, even with that context.

For now it sees things in a mirror, dimly lit (if you will excuse the Biblical reference I got via Star Trek).

> the bloat was arguably already dead weight. but the cut doesnt discriminate, it takes both

It can do, sometimes, but I think it doesn't matter much. The competent developers keep developing just in a new place, while the old products which they no longer work on were often good enough long ago.

Thank you for the correction.

> Bad actors will always exist, but fortunately will always be outnumbered by good actors with access to the same tools. So while I understand your pressing for caution, I still think that your argument is nonsense; bad actors will always find uncensored AI while good actors continue to shackle themselves with censored AI that has failure modes which reduce actual ethical utility. I'm afraid to tell you that the cat is already out of the bag, dude. You're like the guy who wants to leave a sign saying "NO GUNS ALLOWED" just inside a daycare. "Sure, I'll get right on that," says the concealed-carry bad actor...

Guns are an excellent metaphor here, especially as with "good actors with access to the same tools" is a pattern-match to the incorrect statement that "only a good guy with a gun can stop a bad guy with a gun"*. Much of the world outside the USA neither has, nor wants to have, the 2nd amendment. Are gun bans perfect? No, of course not. But the UK (where I grew up) has far fewer homicides as a result, and last I heard when polled on issue even 2/3rds of the UK police feel safe enough to not desire to be armed (though three quarters would agree to carry if ordered).

Similarly, good actors using an AI can only cover the malignant use cases they themselves think of. Famously, the 9/11 attacks were only possible because at the time nobody had considered that anyone might weaponise the vehicles themselves until they saw it happen, which was also why of the four planes only one saw the passengers fighting back to regain control.

In particular, "bad actors will always find uncensored AI" suggests that all AI are equally competent. Right now, they're not all equal, the proprietary models are leading. Of course, even then you may argue that the proprietary models can be convinced to do whatever via the right prompt, and to an extent yes, but only to an extent.

The malicious users can only be slowed down (as opposed to the normal people who simply put too much trust into the current models who can be mostly prevented from harmful courses of action with the same guards). But AI provides competence that bad actors would otherwise not have, so even a simple guard will prevent misuse by nihilistic teenagers whose competence does not yet extend to the level of a local drug dealer let alone the competence of a state-sponsored terrorist cell.

* https://en.wikipedia.org/wiki/Good_guy_with_a_gun#Analysis