To the degree that's true, it's nothing inherent in the processor generations. What's happened is that Moore's Law has slowed dramatically. If there were as much performance difference between a 2026 CPU and a 2020 CPU as there was between a 2006 CPU and a 2000 CPU, we'd still be feeling the need to upgrade regularly.

Consider, for example that between 2000 and 2006, clock speeds tripled, CPUs went from single to dual core, and instructions per clock went up. A 2006 flagship CPU was ~20X faster than a 2000 flagship CPU. From 2020 to 2026 we saw, what, 2X? And most of that gain came from increasing core counts and hybrid big/little cores, which means that most workloads don't realize the full benefit.

When machines are getting an order of magnitude faster every five years, you're going to be upgrading frequently. That hasn't been happening for a while. If we have some major shift in CPU tech that give us 10X faster machines by 2030, the upgrade treadmill will resume.

That's part of what is great about this though. This is happening from sheer economics, despite the Trump admin's attempts otherwise.

omnis tuus basus est belongus ad nos

sigh s/more /

I don't want edit, I want a browser grammar checker that's all client side.

The rates are bad, we don't focus on using the least bad estimate we produce, and we stave off crisis to a degree with mediocre public assistance programs which struggle to cover needs for lack of funding but which really amount to can-kicking. That's better than nothing, but still leaves us poised for disaster. If Cheeto Benito successfully terminates these programs (as he has been trying to do, and he has successfully been interfering with them) then the looming crises become immediate not quite overnight, but in literally more 10-30 days.

Seems like perfect circumstances for using hallucination-prone techniques!

#whatcouldpossiblygowrong

You joke but this is literally how a lot of AI works already. The only thing that is new here is someone discovered it, called it a loop, and thought they were clever. Since the early days of image generation your prompt doesn't get fed to the image generator, it gets fed through an LLM first that creates the scene. Many AI systems already work like this internally.

It's multiple levels of abstraction to get to a resolution.

Customer: I want software which does X.
Engineer: Writes a detailed specification for X.
Programmer: Takes detailed specification and converts it into code.

It would seem like they are replacing the engineer here with an agent. It's also worth noting that precisely none of this is new in AI. In fact this kind of thing has been in the workflows for a long time. Take Nano Banana for instance. If you feed Google's model an instruction they don't just feed it into an image generator. They feed it into an LLM first which creates the scene which then feeds into an image generator.

This is how most of AI works, but someone just came up with a snazzy term for it and thinks they are oh-so-smart.

Today I'm mostly avoiding working on cars, but I do have some ongoing projects and they suck. I can't wait for "every" vehicle to be an EV, which isn't practical for me now, but hopefully will be in the future. But I really don't want everything phoning home with information about me constantly. That information could be used against me, so I don't want it to be collected. Nothing prevents it here except opting out by not buying the whatever-it-is, but sometimes you need the thing.

As vehicles age, they tend to get first cheaper but then more expensive to maintain, so just staying in the past forever isn't realistic. It would be nice to have some options without the constant oversight.

Apparently 93% of the population of the USA does not work for the government, and an even higher percentage not in a critical role. No one gives a flying fuck if Xi Jinping knows where they drive. If your post were to be relevant then the car would be banned for government / critical employees only. Banning it outright has zero to do with foreign surveillance unless you are really really fucking bad at making policy. ... which ... let's face it.

That's a dumb comment. No one would rather die than use AC and the AC use is on the rise everywhere. The problem is houses in Europe are designed for historical European weather, just like houses in Texas are designed for Texan weather. The problem is always a case of adapting. Virtually everyone who is dying because of heat is not in a position to install AC usually for some combination of being unable to afford it, or being unable to get it installed (regulations are a problem, apartment design is a problem).

No sorry equating one and the other even even dumber.

Who? No one in Europe gives a shit what some autistic kid thinks.

Again Spain is not California. A traffic accident is not going to knock out anyone's power in a country where a massive amount of critical infrastructure is protected and underground. Even on a local level the average duration of an outage is 30minutes largely because competent power system design makes it such that you don't need to care if part of your network is taken offline. You just "internet" the problem and route around the broken link. It's why ring topologies are so prevalent on local levels.

Abuse is a strange word which really makes me wonder if you've never met anyone who has actually been abused. This technology has zero meaningful impact on your life. Daddy government, or the Xi Jinping knowing you drive to your mistress when your wife isn't home doesn't impact you. What does impact you is potential local data sharing with companies that have inaction with you directly (e.g. Ford sharing your data with your insurance company).

The technology isn't the problem here. The smarts aren't the problem here. Polestar knowing my journey history isn't a problem here (sidenote: I chose a Polestar because one of the killer features was the automated log-book functionality, not only is Polestar tracking me, but I explicitly want them to and get an excel report of what is on my Polestar account).

The problem is companies using data in appropriately. The USA really needs a GDPR style law, or any kind of consumer data protection laws period.

One of the problem is the use of past tense. It creates a bias. You *worked* on cars and like any good aging nerd you think every change is only a drawback and are completely oblivious to the way normal people interact with cars. https://en.wikipedia.org/wiki/... Even the concept of "working" on your car puts you at odds with the way normal people use a car.

That you have an extinguisher puts you in the top 1% of preparedness. The go-bag maybe in the top 1% of that 1% prepared group. Kudos.

That should have been km^3, obviously.