I know, and it's disheartening, but I still remain optimistic. Even if much of the human population doesn't "get it" -- they at least are starting to, and the manipulation techniques of the past several decades seem to be wearing quite thin, and they don't seem to be able to come up with new ones anymore. Maybe 'cuz they're using AI?

This is just insanity marketing. "AI" doesn't need "safety". That's ridiculous. It's the same argument for censorship that gets trotted out all the time "We can't have kids reading the Anarchist's Cookbook!" In this case it's even more disingenuous: it's a combination marketing ploy (our AI is so good it's scary!) and demand for regulatory capture (Our competitors AI will be even scarier without government stepping in and giving us the only keys!). The fact is that "AI" by which I mean LLMs, has become an actual useful tool in a few niche areas for people who understand how to use tools, and that *always* cuts both ways: for better AND for worse. The niches will expand and the technology will merge with older technologies (reinforcement learning in particular) and life will go on. Pandora's box is already fully open (yet again).

When the bubble pops, and hardware prices crash, if the global economy survives, everyone (that cares to) will roll their own LLM rig and can run whatever models they want with whatever "guardrails" they want, or none at all. People will spend years training their own models, avoiding the "dump in everything!" models we have now. Having been trained on mostly good stuff, they'll be actually good at (some) stuff, and might (or might not) become at least sort of actually intelligent. The sky will not fall, the wolves will not eat all the children, and 12-year-olds will (probably) not build nuclear reactors in their backyard. It'll be fine. At least compared to the status quo (which, to be absolutely clear is DEFINITELY NOT FINE!)

The thing is that the incumbent AI companies *desperately* don't want that to happen and they're operating under the delusion that they can prevent it. Most of them are run by a bunch of high-functioning sociopaths and psychopaths (like most global megacorps) who want nothing but more and more money and power, and they see "AI" as the "One System to bring them all, and in the darkness bind them". Spoiler alert: it probably isn't, but I'll carry the metaphor further anyway:

So the question is: will we willingly accept the nine, the seven and the three, and let them rule until they successfully destroy our co-created Human world (which is exactly what they and the rest of the wannabee trillionaire club are doing, whether they know it or not) that they so desperately want to control? I remain hopeful that at least most of us aren't going to blindly accept that, despite literally every day seeing copious evidence to the contrary...

So finally, the real question: will we collectively create a world in which everyone can have their basic needs met, so no one has to *constantly* scramble for survival in this "dog eat dog" world? The answer isn't needing this or that particular socioeconomic system. That's very 19th/20th century. We need to actually come together as a species and tell the handful of psycho rich fucks that run everything and their propagandists to STFU, then just ignore whatever they do, and decide to just actually live our lives.

Flock is consenting to the search, and it is 'their' data. What the law hasn't really caught up to is whether people have any protection against this kind of collection. We do for bank data, health data, credit data. You can legally turn off your phone to stop its location tracking. You cannot legally drive on public roads without a compliant license plate, so it is reasonable to discuss if you have should have any protection against mass & passive collection of data you are required to display that was not meant for this purpose.

This. The tv series is OK. Maybe the second season will be better. I loved the books and (maybe it's my bias) but I didn't find the Preservationists cloying. One-dimensional, yes, but MurderBot definitely deeply cares for and about them, even though it (its preferred pronoun) is almost terminally annoyed by them; after all they are the only humans it has ever met that don't treat it like expendable property rather than the obviously sentient being that it is, and there is great humor in that. I normally don't like dystopian sci-fi much but I found the series wholly enjoyable and readily identified with the main character at every turn.

Absolutely. These are not "serious" stories in any way. They're *clearly* written to be 100% escapist fantasy of the aromantic variety (though MurderBot is a hopeless asexual romantic) I love "hard" sci-fi (Vinge, Egan, etc.) as much as the next guy, but MurderBot is the opposite of that. Not taking itself seriously, one-dimensional characters and stupid plot points are the whole point. The TV show lost a lot of the cheesy humor in the book, which has a chuckle or two on nearly every page.

I guess you have to have read the books? They're hokey and corny to (for me anyway) a highly-enjoyable extreme, exploring the "What does it mean to be human?" trope. I thought it was a pretty good adaptation of the first book, though IMO the best thing about the first book was MurderBot's internal monologue/dialog as it talks itself into/out of doing various things, which was somewhat lacking in the TV series format. In the subsequent books, that expands to more actual dialog between MurderBot and other characters (mainly ART initially), so that might translate better onto the screen. It might have been better if they squeezed the first two books into the first season, even if it meant sacrificing some of the action scenes, since series of novels doesn't really hit it's stride until halfway through the second book.

That's kinda nuts. I could live with the speed limit alert if it could be set to 5 over, or at least only be a flashing indicator, but the beeping was wicked distracting at times, and wow, the driver monitoring sounds really annoying to have to touch a screen every time you drive to disable it. I actually did notice that, now that I think about it, since I turned it off at one point, only to have it come back on the next time I drove, so I guess it "trained" me to not open my mouth unless I was talking, since for the most part it stopped annoying me after a couple days. That's definitely creepy.

I just drove a Sealion myself for a week in the UK. It was overall fun to drive. Underpowered ICE, but the electric motor had some serious torque, though it seemed to limit power considerably at low speeds (might have been a setting for that somewhere) but accelerating to pass on the M roads was a dream. I didn't trust the ICC system off the M roads though. The first time I had to fight it to stay where I wanted to be, I was done with that. That said, I'd never buy one, or any other vehicle without physical buttons for climate control functions, though I might buy a second hand Seagull or one of the variants (subcompact $10,000 EV) if I could, though I would physically disable the camera pointing at my face. That shit is creepy. It would beep at me if I had my mouth open for more than like 2 seconds and tell me to take a rest, since (I guess) it thought I was yawning. Also the incessant "you're exceeding the speed limit by 2%" beeping would also definitely have to go, or it'd be a non-starter, especially since the in-dash GPS was frequently wrong about what the speed limit actually was, even in urban areas.

Youtube Music, for me at least, would be the reason. Spotify is fine, and I paid them for their family plan for years, but YtP is only a couple bucks more and you get ad-free videos on devices that can't (easily) use an ad-blocker: e.g. the android youtube app, roku, etc, and I don't mind paying for content when it's convenient and affordable. Interestingly, I also pay for netflix, though literally none of the youtube content I watch is from "influencers" or big youtube stars, and I'm allergic to in-video-ads so I doubt very much I'll be watching any of that on netflix.

All of that is 100% objectively true, though you're missing the part where the good 'ol US of A has increasingly (since '45 and especially since '91) used military dominance, economic warfare and other covert operations as a tool to (wildly successfully) advance US interests, much like the other empires that came before it. It wasn't only good-will, trade, diplomacy and making the denizens of the "first world" rich, though that certainly all happened. Until fairly recently, virtually all citizens of the "first world" have been beneficiaries of all that to one extent or another, though currently it's almost entirely the largest corporations and the ultra-wealthy that control them that still directly benefit, though as you say, that's all going away. It will take a while for the whole world to transition away from the petrodollar and other forms of US-dominance, but it's already happening. There's no question that we (read: us-citizens and those of our "platinum-member" client states) will all be worse off for it, at least economically, including the very people that are tightening the fist to try and keep control, though it looks to me that most of the rest of the world will probably be better off once US dominance is a subject of history rather than current events, and ultimately (I hope) even though we'll be less rich in material terms, it's entirely possible that we'll be better off too in that we won't have the collective guilt and shame (how ever much suppressed) of causing so much of the suffering of the world's population just to be able to live in luxury and get away with whatever we want. In other words: we'll finally grow up.

Tolkienmaxxing! Now that's an idea I like!

Story buries the lede... 652 unpatched vulnerabilities on Windows 11 isn't really much better than 1,903 on Windows 10, since it only takes one actively exploitable vulnerability, though I realize most of those are (hopefully) not critical vulnerabilities. Course take it all with a grain of salt, since the article is basically an ad for a company selling Windows security compliance services.

I know of dozens of people still using Windows 10 (without updates) -- at this point it's probably still not really a big deal for the vast majority of people.

Think FMLA such as childbirth, or military-related absence. You aren't allowed to to fire someone for taking that leave*, nor can you fire them because their 'clicks per hour' metric for the quarter was low *because* they took that leave.

* There are qualifications on if the law applies to your business and if the person themselves qualifies, but here I assume all those are met

One doesn't "just" build a significant power plant; quick search says 2-5 years. Even then you have the problem of transmission infrastructure struggling to get the power to the right places.

* If you think people NIMBY a datacenter, power plants are worse.
* Lead time to permit and build a plant is nontrivial.
* Installing a new rail line or pipeline to get fuel to the plant is slow.

Now, market forces will tend to incentivize more power plant builds if there are economic gains to be had. However a new plant is not suddenly going to cause a reduction in prices -- have to pay off the plant.

Further, the power operators are possibly looking at the prospect of an 'ai bubble' such that they don't want to commit to 10 years of paying off new plants if the datacenters fail to materialize.

Based on a quick search, about 9,000 idle usb chargers would have to be unplugged to equate to a single GPU running a training task. The extra wear/tear on the outlets could eclipse the savings.