This is why the more I think about it, the less I’m afraid of AGI. I mean, if the super basic non-thinking AI we have now, developed by capitalists and programmed to be loyal to the bourgeoisie, ends up like this.

Really seems like Deepseek is one of the only vendors actually focusing on performance per unit compute power and not just throwing infinite compute power at the problem. Calling it now, when the bubble bursts they’ll be one of the few to make it out with a usable product.

Why does Ollama only have a cloud version?

People can run it to verify? That’s how open source models work.

Gonna link to this thread the next time someone gives me shit for adding a /s to what they thought was “obvious” sarcasm.

Fucking cowards using flying drones. Kill me with a full humanoid and then I’ll be impressed.

There are hearing aids that look identical to wireless earbuds. I sometimes wonder if people using them get judged because everyone assumes they’re talking to them with earbuds in.

That’s fair. That’s honestly the optimal use case. But with how fast technology tends to advance I suspect this will get marketed as “everyone should get one and you’re basically Amish if you don’t” before you know it (even if China’s socialist government has regulations on how companies can market stuff, we Westerners don’t and it’s a matter of time until some Western techbro company contracts this into the next gadget people will judge you for not having.)

People already judge me for being in the tech industry and refusing to use whatever latest app/service that’s going viral because I know how easily untrusted software can fuck you over.

Look, I get that this is impressive and all, but as a developer, keep that shit out of my brain.

And this is not a “hurr durr China dictatorship hurr durr SeeSeePee thought control” thing, it’s an “I have no reason to think they don’t have good intentions but software, no matter how genuinely well intentioned, will NEVER be trustworthy enough for me to connect my brain to” thing. We DO NOT understand the brain nearly enough to even begin to theorise what could go wrong or what kinds of exploits could be done on it. Maybe it’s because it’s a paradigm shift that I’m not prepared for, but this sounds like absolute security hell and can only go disastrously wrong. And you can’t wipe and reinstall your brain when it does, nor does evolution tend to patch CVEs in a timely manner.

Remember that there are ingestible chemicals, a comparatively extremely crude and imprecise way of controlling the brain, that can completely destroy your mental defenses and make you do whatever the attacker wants. Imagine a direct neuron level interface with a computer that can scale infinitely beyond the processing power of the brain. Fuck that shit.

OK but the whole point of math is once a proof is articulated, it is trivial to verify its correctness. It could have come from the devil himself but if it checks out, it’s as valid as any other.

Whether the AI is “actually thinking” is irrelevant. We’re thinking, and we can verify proofs objectively. People say AI art isn’t real art and there are definitely arguments for that, but saying AI math isn’t real math is kind of ridiculous.

Because it’s the only Canadian region with decent rail transit /s

Just block and move on

Why haven’t you? Take your own advice.

The AI started writing beyond their primary audience’s reading range /s

These are the same companies who insist on kernel mode anticheat

Something tells me it was never about “the integrity of the game.”

First time I’ve seen CAMM memory in a real product. Pretty cool, but not sure if I prefer it to traditional DIMM slots though.

But the removable ports are a fucking godsend. So sick of a broken port making an entire motherboard unusable.

Honestly the only thing more I can ask for is for the battery to be on the outside of the case (like old school laptops) so you can replace it without opening it which not everyone is comfortable with doing. Otherwise this is really good. The only thing stopping me from buying this laptop is the fact that my current 6 year old Lenovo still works perfectly after I replaced the battery (honestly Thinkpads have always been pretty repairable, this is going above and beyond).

What happened? Were they caught altering archives?

Doesn’t simp for the US, just coincidentally believes every claim the US makes about socialist countries. Cool.

Post your sources then. I posted mine after I “got caught in a LIE.” Since you’re obviously not LYING and are a well-read expert on Cuban labour practices unlike me, that should be an easy ask for you.

https://github.com/dessalines/essays/blob/main/socialism_faq.md#on-cuba

Not a source in itself but links to many sources.

Also, you know nothing about Cuba and everything you say is straight out of the US propaganda pool which they’ve been using to justify their illegal blockade of the country. Also also, if I’m just a hive mind drone and you’re apparently not, why not post your own sources to show how much better informed you are and how much better you are than me? Seems like you don’t have sources and are deflecting from your lack of sources.

That, if true (since you didn’t link any sources), sounds at worst like the kind of exploitation that Westerners like you and me rely on (from “undocumented” or even documented immigrants in your own country, resource extraction sectors in developing countries, cheap overseas manufacturing, etc) that allows us to make the illusion of a living wage from what little work we do in comparison. In reality Cuba would actually be far less exploitative than us because we do all of what you described and much, much worse to people you never see or think about. Where did that cobalt in your phone come from? What about your chocolate? Where does your e-waste go after you drop it off at a big box store before buying the newest thing? Cuba at least doesn’t prop up their own citizens at the great expense of people they deem less important.

Cuba for one. China has been working very hard towards it (to a much further extent than any Western country) but due to their massive population it’s debatable of they’ve gotten there for every single labour job yet, but they will.

Ones that don’t exploit immigration or prison as a source of cheap labour. AKA nowhere in the West.

“China has exploitative labour!”

China: “OK we’re using robots to keep labour costs low without exploiting anyone”

“Wait, not like that”

Isn’t OpenAI also investing heavily into Nvidia? How the hell does that work? They’re just circulating money around?

Don’t worry, one look at you and they’ll know you’re not worth harvesting.

Deadlock will not be a MOBA on the condition that Deadlock is not a MOBA

These are copying human movements i think

You’re saying that like it’s not impressive even if that were true. Human motion capture can’t react fast enough to loss of balance with the speed of the moves they were doing (neither can a fully pre-programmed routine with no real time processing), especially for something smaller than the human operator for which you wouldn’t have the intuition of how to right it, especially with conventional motion capture that doesn’t give you feedback like moving your body does (unless they built a fully featured motion capture system that’s both super low latency and gives the operator complete and accurate mechanical feedback on all degrees of freedom, in which case you’d think they’d feature that more than the robots themselves because that’s way more impressive.) Even if they were all piloted by humans, they’re still autonomously processing all those motion inputs to be compatible with its body.

But yeah, there is an attempt to make these robots look cool so humans wont resist when they patrol the streets.

Don’t worry, we’ll never have this problem in the West because humans over here like being police bastards too much to let some robots take their fun away.

OpenAI: “They stole our technology!”

Also OpenAI: “Uh, well, our technology is actually inferior to theirs, but they must have stole it and made massive sweeping improvements to it that we weren’t able to! How dare they!”

The Elon aspect has already been covered by other commenters, so purely against Starlink’s technology as a primary method for internet, ignoring Musk:

It has an enormous carbon footprint. Launching stuff into space takes a ton of energy, and SpaceX rockets are entirely powered by fossil fuels. Most of the rocket body is just massive tanks of fossil fuels, and because they don’t fly very far from Earth, most of that ends up in the atmosphere. The internet already has a significant carbon footprint, and adding this layer when we absolutely don’t have to is stupid. We can build A LOT of terrestrial radio infrastructure for less environmental impact, covering pretty much all rural areas. Microwave dishes pointing to towers is superior for rural internet in pretty much every way, including latency which is Starlink’s main selling point over older satellite internet systems, and wired internet is still the best option in every benchmark possible so using Starlink in urban places where you can effectively supply wired internet is stupid.

But what about people who live in super remote areas where ground based infrastructure is unfeasible? Well, we’ve already had internet capable satellites for much longer, and Starlink is an inferior satellite technology in terms of efficiency compared to satellites that orbit much higher up. They fly so low that most of the time they’re doing nothing because they’re flying over the ocean or places no one is using the service. With geostationary satellites, each satellite can “see” a larger portion of the Earth, so not only do you need fewer satellites while still providing global coverage, each satellite is in use much more of the time even when they’re flying over unpopulated areas because they cover so much more area, so say, ships and wildlife researchers in the jungle can stay connected to a single satellite instead of needing a dense web of satellites flying by overhead to deliver continuous coverage.

Flying so low also causes them to experience much more atmospheric drag, meaning they have a much shorter life. So you need more launches in total to replace satellites and maintain global coverage, massively increasing the carbon footprint. You also further pollute the atmosphere with vaporized satellites (which contain some nasty heavy metals BTW) when they run out of propellant and fall back to Earth. So not only do you need fewer satellites with geostationary orbit, each satellite also has a longer life.

The antenna you’d need on the ground is also much simpler, just a dish instead of an expensive, fragile, and power hungry phased array. Pretty important for truly off grid people.

It’s also bad for national security (again, speaking on national security implications of the technology in general because as a Canadian I couldn’t care less about US national security) to rely on it as a primary way of getting Internet because, as we’ve just learned, other countries can just shoot down your satellites when they fly over their territory. Not helped by the fact that they’re so close to the ground. It would be a lot harder to attack infrastructure in a country’s own territory. And if you’re not the country operating it, you’re also at the mercy of that country because they can just deny you access.

(Edit: made a more formal comment closer to the root of the thread)

Why? Launching shit into space is hard as fuck and has an enormous carbon footprint. You can build A LOT of cellular infrastructure for the same cost and impact.

And building your internet infrastructure in your own territory instead of floating in space will make it a lot harder for China to shoot with their badass microwave canon.

And I’m just a common idiot, but I’d wager upgrading satellite infrastructure is going to be slightly more expensive than terrestrial infrastructure. There’s a reason we’re still using a lot of satellite infrastructure from the 1980s.

his Asperger’s

As an autistic person, don’t you FUCKING DARE defend his use of that fucking card for the horrible way he acts.

Same applies to all social networks and MS Windows, right

YES

What about cars?

/c/fuckcars

Just wondering how far this govt coercion hatred extends into everyday life.

A lot

Which is also fucking horrible. Yeah just further pollute the atmosphere with vaporized satellites why don’t you. Hope that Fortnite battle was worth it.

Just one of the reasons removable batteries need to come back. Imagine if you could just choose between the lighter lithium ion and the more durable lithium iron phosphate for your use case.

LiFePO4 battery is a no brainer for cars. Higher current, longer lifespan, and much lower chance of turning into a bomb in a crash.

A common argument I hear against using it in mobile electronics is the fact that it’s heavier. But, like, who cares? Are you struggling to lift your phone or laptop?

Can we at least wait until we develop AGI before we start the robots enslaving humans stage? I’d like my slaver to know what they’re doing, thanks. /s

In other news, Tesla to introduce buttonless handless doors. Tell Grok why you deserve to be let out.

It always blows my mind just how much resources companies are willing to spend on DRM. Like, surely at some point your R&D costs will outweigh whatever piracy you might have prevented, and that prevention rate will never, ever be 100%. And that’s assuming they spent extra resources on DRM and didn’t take it out of the actual game development budget, resulting in a shittier product and less sales as a result.

It reminds me of when the transit system in my city introduced fare gates. It massively inflated the operating cost and guess what? It only ever stopped honest people who either forgot to load their card or were new to the transit system/city and didn’t understand the zone system, so loaded a 1 zone pass and had the audacity to ride even one station outside the city they got on (not to mention when the system glitched and refused to let you out even when you did pay). The people habitually not paying just casually push past the fare gates and no one stops them. I’d genuinely be suprised if they’re even breaking even with the operating/maintenance costs vs whatever few unintentional fare dodgers they manage to stop. Most likely they’re losing money, while making the transit system less efficient by introducing a bottleneck, while discouraging drivers from trying out transit, just because they can’t stand the idea that people can just walk on the train without paying (even though they haven’t actually stopped them).

But won’t you think about the silicon fab duopoly? They are the true victims in this!

Surely the US won’t freak about this like massive hypocrites, right?

So, instead of feeding large documents into these models which break them, you can instead provide them with an API to interrogate the document by writing code

Kind of off topic, but this reminded me about something I really don’t like about the current paradigm of “intelligence” and “knowledge” being parts of a single monolithic model.

Why aren’t we training models on how to search any generic dataset for information, find patterns, draw conclusions, etc, rather than baking the knowledge itself into the model? 8 or so GB of pure abstract reasoning strategies would probably be way more intelligent and efficient than even a much larger model we have now. Imagine if you can just give it an arbitrarily sized database whose content you control, which you can then fill with the highest quality, ethically obtained, human expert moderated data complete with attributions to original creators, and have it base all its decisions from that. It would even be able to cite what it used with identifiers in the database, which can then be manually verified. You get a concrete foundation of where it’s getting its information from, and you only need to load what it currently needs into memory, whereas right now you have to load all the AI’s “knowledge,” relevant or not, into your precious and limited RAM. You would also be able to update the individual data separately from the model itself, and have it produce updated results from the new data. That would actually be what I consider an artificial “intelligence” and not a fancy statistical prediction mechanism.