If you gave your AI permission to run console commands without check or verification, then you did in fact give it permission to delete everything.

A lot of them been so indoctrinated into mistrusting authorities and instutions, that they basically disbelieve anything they say on principle.

And al the evidence, all the scientists telling them they’re wrong just ends up reinforcing their belief in some giant conspiracy.

It’s sadly been shown in more than one study that changing the mind of conspiracy theorists with reason, argumens or evidence is basically impossible. It’s almost a self preservation instict against cognitive dissonance. They were so sure they were right, and now so one is telling them they’re not. That feels shit, and it feels shit to accept you were wrong about something you so fervently insisted was true. So their brains basically go into self defense mode, and just reject and attack anything that threatens the shaky fundamentals of their entire belief system. The best thing you can attmept to do is to distract them. Get them to talk and think about other things. When they mention the conspiracy, don’t engage, don’t argue how they’re wrong, they’ll just dig their heels in deeper, just change the topic to something else. Force them to spend less time in their delusions. Eventually, if you’re lucky, they might gain enough distance to the topic, and stop caring about it enough that they’re ready to start accepting how batshit insane those conspiracies are.

To quote Hbomberguy:

“The anti vaccine movement had next to no evidence [27] years ago, and they have even less today”

That is enormously ironic, since I literally never claimed you said anything except for what you did: Namely, that synthetic data is enough to train models.

According to you, they should be able to just generate synthetic training data purely with the previous model, and then use that to train the next generation.

LIterally, the very next sentence starts with the words “Then why”, which clearly and explicitly means I’m no longer indirectly quoting you Everything else in my comment is quite explicitly my own thoughts on the matter, and why I disagree with that statment, so in actual fact, you’re the one making up shit I never said.

Ok, that is a fair point I hadn’t previosuly considered. Though disclaiming a patent doesn’t loose you all legal recourse.

If someone else tries to repatent it, even if it gets approved, you can still file a challenge against the new patent with the PTO. You (or anyone else, really) would also have a virtually guaranteed court win, even if someone got the patent through and tried to enforce it. All you’d have to prove in court is that prior art of the invention exists, therefore the patent is invalid and unenforceable, granted or not, so it’s unlikely someone would even bother trying to enforce such a patent. A previous, diclaimed patent, of literally the identical technology being on record is pretty iron clad and unavoidable evidence that the patent isn’t original.

Yeah, it is a downside that this may discourage EV adoption, but tbh. if you chose your car based on the fines you’d get when speeding is…questionable, to say the least.

“the consequences of doing an illegal and dangerous thing with my product” should, in my opinion, not be a choice criterion.

Though a scaling that takes both size and weight into account might be even better. Particularly for crashes involving pedestrians, cyclists and bikers, the size, and particulalry the height of the bonnet, are almost more significant than just the raw weight. A hollowed out F-450 weighing 1 ton is probably still more deadly to those people than a ~1.5-2 ton electric sedan.

This isn’t really aimed at reducing adoption of big cars, more at getting the people who drive more dangerous vehicles, to drive with more regard for safety, and be penalised heavier when they don’t.

Most common fission reactions today release most of their energy in the form of neutrons. The only way to extract energy from neutrons is heat. But there are fission reactions which release a large portion of their energy in the form protons. And since protons are charged, their energy can be electromagnetically converted directly into electricity, with no need for intermediate process steps.

There’s already at least one company building prototypes like this, Helion, using D+He3 fusion, rather than the more common D+T fusion in other reactortypes like Tokamaks.

Real engineering has a video on Helion: https://www.youtube.com/watch?v=_bDXXWQxK38

If the model collapse theory weren’t true, then why do LLMs need to scrape so much data from the internet for training ?

According to you, they should be able to just generate synthetic training data purely with the previous model, and then use that to train the next generation.

So why is there even a need for human input at all then ? Why are all LLM companies fighting tooth and nail against their data scraping being restricted, if real human data is in fact so unnecessary for model training, and they could just generate their own synthetic training data instead ?

You can stop models from deteriorating without new data, and you can even train them with synthetic data, but that still requires the synthetic data to either be modelled, or filtered by humans to ensure its quality. If you just take a million random chatGPT outputs, with no human filtering whatsoever, and use those to retrain the chatGPT model, and then repeat that over and over again, eventually the model will turn to shit. Each iteration some of the random tweaks chatGPT makes to their output are going to produce some low quality outputs, which are now presented to the new training model as a target to achieve, so the new model learns that the quality of this type of bad output is actually higher, which makes it more likely for it to reappear in the next set of synthetic data.

And if you turn of the random tweaks, the model may not deteriorate, but it also won’t improve, because effectively no new data is being generated.

I mean they’re not mutally exclusive. Fines could, and should scale with both.

I’ve always said that the fines for moving traffic violations need to be scaled to the weight of the vehicle.

It’s insane that going 20 over the limit on a 250kg moped gets you the exact same fine as doing it in a 4 ton F-450.

I was referring to the effect being opposite of the effect that would have been motivated by capitalism.

This is a single company making a decision. It is neither capitalism nor socialism nor any other economic system, because it’s not a system. It’s one company making one decision. I just assumed that was obvious from context.

And my point was that this decision is quite literally the opposite of the decision that capitalist incentives would drive the company towards.

And yes, government regulations could accurately be described as the opposite, or in direct opposition to, free market capitalism.

What the hell even is the point mandating a back up alarm for self driving cars ? Backup alarms literally only exist because visibility to the rear is worse, and to warn pedestrians that a vehicle nearby is moving with very poor to no visibility, but that only applies to human operated vehicles. Autonomous vehicles use 360° sensors, they can “see” just as well in reverse as in forward. Be that good or bad, it’s equal in every direction, so mandating an alarm just for reverse seems enormously pointless. Especially since the cars tend to be slower in reverse, so if anything it’s less necessary then, vs. when they’re moving forward.

The line, imo, is: are you creating it yourself, and just using AI to help you make it faster/more convenient, or is AI the primary thing that is creating your content in the first place.

Using AI for convenience is absolutely valid imo, I routinely use chatGPT to do things like debugging code I wrote, or rewriting data sets in different formats, instead of doing to by hand, or using it for more complex search and replace jobs, if I can’t be fucked to figure out a regex to cover it.

For these kind of jobs, I think AI is a great tool.

More simply said, I personally generally use AI for small subtasks that I am entirely capable of doing myself, but are annoying/boring/repetitive/time consuming to do by hand.

There’s two sides to this, both of which are equally true.

1. Cars are a fuck ton more dangerous than most people realise/think

2. Violent crime is a whole lot LESS common/of a risk than sensationalist media makes most people assume.

Nominally you can use it to plug a generators output into a household circuit, which will provide power to that circuit in cases of a blackout, saving you from needing to unplug everything critical and daisy chain 10 multiplugs to the generator.

It could also be used to connect two seperate household circuits together, if only of them is actually live for whatever reason.

In reality you shouldn’t his at all, ever. Just daisy chain the extension cords. If you forgot to isolate the circuit by flipping the main breaker (easy to do if there’s no power anyway, because of a blackout), and then the grid comes back on, your generator is gonna have a real bad time. And then there’s obviosuly the electric shock risk of using something like this.

Ticket Master is a massive monopoly in the US, and has exclusivity contracts with most of the popular concert venues, meaning if you want to perform anywhere of even mediocre popularity/size, you have literally no choice as an artist.

If “everyone will be using AI”, AI will turn to shit.

They can’t create originality, they’re only recycling and recontextualising existing information. But if you recycle and recontextualise the same information over and over again, it keeps degrading more and more.

It’s ironic that the very people who advocate for AI everywhere, fail to realise just how dependent the quality of AI content is on having real, human generated content to input to train the model.

Not to necessarily defend the idea in the article, but that comment screams that you just read the headline and not the article.

If you had read the article, you would know that the author doesn’t want to get rid of routable addresses, they want to replace the current system of IP address assignments with an automated cryptographic address system, allowing network size to rapidly increase, and self organise without reliance on a central address authority. So your analogy of having no address at all is massive misrepresentation of the authors idea.

Wildly misrepresentating ideas is never good. Even if you dislike it, by wildly misrepresentating the idea, it just discredits your own stance, because it’s (seemingly) based on falsehoods.

Pretending like the author just wants to just abolish all types of routing addresses is dishonest.

Except, Airbus fixing this has fuck all to do with capitalism. They are legally mandated too. If it was Worker owned, they’d still fix it. If they were state owned, they’d still fix it. If anything, the actual forces and incentives that are inherent to capitalism, and not just existing alongside and in opposition to it, would drive them to not fix it.

Fixing it has literally nothing whatsoever to do with Airbus being capitalist or profit driven.

Just because something happened in a capitalist system doesn’t mean it happened because of the capitalist system. That’s kind of like saying capitalism caused workers rights, because companies nowadays have an incentive to not get sued for worker rights violations, which is obviously nonsense.

If you’re this bent on defending this mysoginstic and sexist crap, then you’re a sexist mysoginst not worth talking too. Enjoy being blocked.