Any rocket or bullet turned against Israel for the past 20 years has its roots in Iran. Iran controlled Lebanon, Syria, Yemen, Gaza and Iraq through its militias.
I’m totally guessing, but I can’t imagine CUDA has any significance at the frontier lab scale. The operational and capex costs are so massive that the convenience of the platform becomes a minuscule consideration.
It’s just that Nvidia’s stuff works, and available at scale, and includes the full stack with networking, cooling and such.
This is all just running in circles. The models are not obviously better. The pricing fluctuates or offset by some other less-obvious metrics (availability/speed/tokens per task/dumbing down). Everyone reports different outcomes in their usage because it's all so context and user dependent. Sometimes models do some things better but become so annoying and obtuse in their other doings that it's just not worth it (like Opus with the insane code comments and Astra with its over-the-top, everything-is-a-sales-pitch style). It feels like the AI gods just turn the knobs on things like compute to get the results they want to align with the IPO to make headlines.
They are pushing their customers towards Chinese models and providers. If you want to get something cutting edge done in defense, cyber, biology - something that isn't common knowledge - you need to venture east. That's an incredible side effect which the Chinese government surely enjoys.
It makes a lot of sense. I thought about it in the context of pull requests or change sets: if the text-to-code process is reliable, why don't you give me prompts instead of code? Code becomes just an intermediate representation.
It's a combination of a park, a residential neighborhood and a shopping center in the renovated power station. The station is incredible building, with its brick construction. The residential buildings themselves are also special, with some designed by Frank Gehry, and the shopping center is inside the power station building and has an family-rated-steampunk feeling to it.
We stayed in an apartment overlooking the station front and it was a lovely sight.
Come on now, when the rich go over to the artsy neighborhoods and fancy wine shops open up it's called "gentrification", but when they - actually a Malaysian group of investors - make their own place in a run down corner of London, a place that's accessible by public transport to anyone, and does justice to preserve the story of the power station, you call it grotesque.
I beg to differ. I stayed in Battersea for a week while exploring London. I found the place very welcoming, safe and inspiring. It’s convenient to get to and from with its dedicated underground line.
I can see how it fits the Apple
brand experience very well (they don’t do gritty stuff). In fact I believe there’s already an Apple Store there.
Programming has a long standing culture of accepting the code to be somewhat wrong, so we have various tests, linters, reviews and error handling. Also in programming there are many ways to do something and it's the end result that matters most.
Not so in other knowledge work. There's no test harness for a contract and error is non-recoverable. Likewise in finance. There are specific ways of doing things and these ways are many times set in regulations. LLMs can assist all day, sure. But replacing the human, in highly regulated, zero tolerance for error environment?
Sorry what claim again needs any evidence here? This is all common knowledge. Iran supporting Russia is a fact no one tries to hide; Iran supporting Hezbollah in Lebanon, Houthis in Yemen, various Shiaa militias in Iraq, supported Assad regime in Syria and Hamas in Gaza - that’s all very public. By “support” I mean that Iran provides them training, weapons, money and direction.
You’re right; given that most of the money in the AI market is injected through OpenAI and Anthropic (which collect it through both selling equity and through customer revenue), the 7-8T is just a derivative of that.
It can't hallucinate, but it doesn't mean it can't make wrong decisions. Just because it adheres to a specific output format at all time, while LLMs have the output format at their mercy, then the claim of not hallucinating is made technically true.
I think that this specific part is not super interesting if your harness just recovers from invalid LLM outputs.
The latency and cost - yes, those are super interesting.
We have an agentic system that produces insights for end users, and runs most of its work on DeepSeek v4.1 Flash but as an output stage transforms the resulting text through Gemini 3.8 Flash for readability, and it works.
On my TODO is try and run all of the analysis pipeline in dense "machine speak" to save on tokens and just let Gemini sort it out at the end.
Reading your comment, I asked it to count how many characters are in the word. It answered correctly: 9. It then also said it skipped an E in the prior chat and corrected it count to 4. It spun for 52 seconds so maybe some Python was involved.
In the savanna it’s not about who outruns the lion but who outruns their peers escaping the lion.
Frontier labs need these headlines not for human progress but as beauty pageant for investors and government agencies. If they don’t do maths they’ll just go after other fields.
So Terrance Tao here might be able to hold them off math but he won’t stop them from speedrunning STEM with similar consequences.
We may be locking people out of these fields instead delegating everything to machines, and I don’t think the machines are good enough to assume that responsibility.
Code cost is almost down to zero. If you move the point of “just leave it to the machine” from the compiler (where humans used to do the coding) to the high level logic (now with LLMs) then in most cases more code does not really matter. Like, why build and maintain an abstraction where the LLM could implement this many times over each time with different subtleties? Why use a library with its own constraints when you could have exactly what you want? Why use cross platform frameworks when you can just one shot the thing to N different platforms?
It’s not even slower. You can have code that’s larger yet more performant (stripping away abstractions can do that).
From time to time I try to do a pass of coalescing flows and cases and removing dead code to reduce the context and prevent the LLM from tripping over itself. But if it’s exclusively LLM maintained code I don’t care too much if there’s more of it.
As a pretty avid Star Trek watcher from childhood, I found most of the things there technologically plausible other than the conversational nature of the ship's computer. Well, LLMs now are far more impressive conversation counterparts than those ships were ever depicted.
I find I still add value, but I don't know if my value is real. Am I just biased and expect things to be done a certain way and penalize the model for doing something different? and am I providing the model with enough high quality context to align with my expectations in one-shot?
I share the sentiment but the question here is how long can we maintain the balance point where the human in the agentic loop is required. It might be a window lasting only a few years, or for the foreseeable future; I think the answer lies the opaque compute economics of the frontier lab: how well models keep scaling and how economically sustainable is serving those models under the current market conditions.
These inventions all stopped disrupting the world and just became a part of it. The question is whether LLMs are going to just take their quiet place, or profoundly change (or eliminate) humanity in a self-feeding frenzy towards singularity.
We need purpose, as a species; after all we left the caves for a better life. Desires make us human, the pursuit keep us busy and entertained.
There is no passion without wanting, and without passion what is the reason for living?
People need purpose. Workplaces provide a very structured, highly managed, perpetual treadmill of purpose. Not purpose in the "self growth" sense, but just something to keep people occupied. "The devil looks for idle hands", and with a large swath of the population* on welfare, there's a good chance he'll find plenty.
[0] that said, population in developed countries is naturally declining so maybe it won't be a problem in the long term either; or maybe, with less need to work, people will be more inclined to raise other people, leading us to both more machines and more humans.