I’m sick and tired of the Capitalist religion and all their fanatic believers. The Western right-wing population are the most propagandized and harmful people on Earth.
Sounds like standard 3-letter agency behavior. Find a known page with some nasty on, store on archive, complain about all the ‘childporn’ on archive. Standard playbook, but they may have gotten some easily propagandized fanatic to act as the moron complainer… Time to support archive.** against the 3-letter psychopaths in the US…
I think US are now ‘nanoseconds’ AFTER China, and that gap is just accelerating faster and faster. US/western commoners should prepare for when the AI bubble-financing implode…
I certainly want it. The current anti-AI wave is irrational.
That isn’t to say that AI is not a major potential existential problem when it goes to AGI and to ASI, or that AI in the hands of private Oligarch psychos are not a catastrophe, but the current hatred is of a different kind, are off the charts, actually evil, and have no relation to the tool it self, or the potential existential risks. The group have already been consumed by hatred and unfortunately, they are now also attacking ordinary Lemmy users that ‘admit’ that they use AI, or ‘admit’ that they treat it as a ‘friend’. You have all seen this. People just mentioning AI are downvoted and hated upon, people that use AI always get a “AI slop !” remark with them on their way. A large percentage, like this post, are subtly attacking that others even have a choice of utilizing AI in their daily life - because they hate it. This is kinder-garden level bullying !
Most of the daily worries of AI, stems from the societal system we are in. Job loss (It’s build into the current system!), ‘AI’-slop (as if Capitalism didn’t create slop?), ‘Copyright’ idiocy that are only necessary in a combative type of society, AI+biz out-competing ordinary people, becoming poor, homeless etc, big monetary/political powers consolidating more power by misusing AI, an incredible amount of propaganda/marketing and manipulation and so on. The old system is incompatible with AI, so people are seeing things go in a wrong way, and feeling the cracks and dissonance between an old failing system, and the new possibilities we have. We - humanity - are in a transitional phase, where the old systems we follow, are no longer adequate, and will implode/convulse in to a new stable system. But the old system is stubbornly trying to keep going, while not having any answers, or any will, to adapt or even discuss the internal dissonance that are accelerating. THAT is the problem! Almost all current ‘AI’ problems that result in this hate backlash - comes directly from the failing type of society we have - not the tool/technology it self ! People should focus on the root-problem, not the tool it-self, and certainly not other users that need it.
This whole group of angry people need to chill tf out and admit to them selves that they are mis-directing their hatred (valid or not), and that their hate are currently harming other ordinary people - not much different from them selves. Stop bullying others just because you can’t handle the inherent failings of the old crap system, a system build on pure combative ideology, “by the oligarchs, for the oligarchs”. Stop fighting for ‘Jobs’, and start fighting for a system where we don’t need ‘Jobs’, Stop fighting for ‘Copyright’, or any of the other insane protectionist ‘patches’ to the current bunkers society/ideology, and start fighting for a society where we don’t have to hoard ‘property’ to survive. Fight for a society that are compatible with where we are technologically/philosophically, one that are actually designed/intended to be compatible with peoples needs and their humble wishes for a safe and stimulating future.
Lastly, being worried/focused on an ASI takeover of humanity, or common people being (more) controlled by the Oligarchs/Big Corp, etc, is completely legit, and it absolutely needs to be talked about, but this group of haters is just hating ‘AI’ and people using it in general, for the wrong reasons, and then lashing out towards those people that enjoy/need it now. (Good for you if you don’t need it in this society).
This bullying has to stop. In the mean time, I and many others, are starting to use current AI to design/build systems that bring us all over this ideological molasses we are in, and land us on the other side as fast and safely as we can. You should probably do the same instead of being consumed by anger, but that is your choice to make. Thanks for reading.
I have not tested it, but Zluda are are drop-in cuda replacement for non-nvidia gpus. The speed should be great. You could check if your goto card is supported…
A backup drive ? dotfiles (.config) ? /tmp /var/log /boot ? map it into your home for something temporary - Downloads ? Slow audio files ? If you like to tinker for fun: btrfs / bcachefs could perhaps utilize it for something, like automatically offloading unused/slow files to the slow usb ? Raid ? ;) Create a partition and let zram swap compressed unused memory pages to the usb partition (I think it keeps pages compressed now ?). You can also run ‘usb over ethernet’, so another machine can use it directly - not sure why tho ;)
On a side note. I still use one of my first usb 1.? drives with 32mb in my old router. 32mb is still 12mb more than my first hard drive, and it fits well with the small openwrt packages. It even has a little slide button to select ‘floppy drive’ mode and a physical ‘readonly’ mode. I wonder what the smallest/oldest usb stick people still use, are ?
Looks cool and simple. Here’s for convenience: https://www.youtube.com/watch?v=IkriZGyjoeU
Agree. “Are you perhaps too angry?”, “The HELL I’m not! F*ck off !”, “alrighty then”…
“\C-x(”: “s/\([[:space:]]\)\(\(\([[({]\(\([[({]\(\([[({]\([][(){}]\{0,1\}\)*[]})]\)\{0,1\}[][(){}]\{0,1\}\)[]})]\)\{0,1\}[][(){}]\{0,1\}\)*[]})]\)\{0,1\}[][(){}]\{0,1\}\)*\)/\1(\2)/n”
YES! Finally something with a manly learning curve ! Enough with all this baby high level programming languages and UI’s. ALL computer interaction should preferably be in a random mix of brainfuck and ed(2) ! Maybe with a little regex expressions randomly inserted here and there to keep us sharp ;-)
Not judging ? I assume you mean ‘free’ ? It is more difficult now. Groq.com once claimed that they didn’t collect data from the free service, but I just checked their privacy rules, and they have now learned the proper Capitalist virtue of surveilling their users and sell their data. Perhaps one of the other fast npu endpoints, cerberus, or sambanova, that are not directly focused on selling data ?
Duckduckgo ai doesn’t require login, but are very limited for chat.
Alternatively, pick free Chinese AI services ? I guess the data they can accumulate are pretty useless to them unless you buy much over there, but not sure to be honest.
Maybe a vpn and gpt4free ? Someone can see and train on your messages alright, but they don’t know who. This gives different models each time, so maybe not cool.
Something like aihorde maybe okay. Your messages are openly processed elsewhere, but you are not likely to be tracked. There’s a lot of personal chat models available. It afaik still requires a ‘small’ nvidia card to join in. Vpn could help anonymity.
For those that don’t mind local AI, there’s a setting that uses localhost:8080 as ai endpoint. You can throw open-webui, or ollama/others at that port.
Compute prices are not going down. If you want a budget friendly compute, then you may need to look for NPU’s. They are slower, and is unfortunately also getting more expensive pr TOPS. Electricity prices in the West are going up as a low output gets bought up by datacenters instead. The newest GPU’s are more efficient in re. to power pr Top, but they are not budget friendly.
I have not seen a recent comparison of GPU vs Npu; TOPS vs price vs power consumption, but @ aliexpress, I saw a small 40 TOPS Npu as a nvme stick with 16gb ram that draws 10 watts, or so (search for ‘ai accelerator nvme’). This little thing can scan files 24/7 and your i5 can help out in peaks. Afaik you can run a distributed model that also runs in, and uses your i5 memory/compute, so if you max out i5 memory, perhaps the combined compute is enough for a larger model ? Maybe a few Npu sticks can work together on the same model ?
Alternatively, you could look for the next version of a Huawei gpu card (first had baby kinks afaik), or one of the other upcoming producers from China. They’ll come faster and faster, but are earmarked for local consumption first.
Another suggestion is to buy one of the old P40/80 (from the ‘pascal’ chipset, I think) (or was it k40/80 ??). They should still support a descent range of modern quantization needs and often have 24gb ram. A refurbished miner card cost around 50-70$+ - cant remember exactly tho.
Lastly, you could try something different. If the i5, with enough memory, can run a large model slowly, you could add a dedicated KV cache, so most of the tokens won’t need to be recalculated. Memory and bandwidth are the most important here, but any old server could be upgraded to be a dedicated KV cache server (might need a network upgrade tho!).
Anyway, Ideas are easy and cheap. If I were you, I would ask an AI for a little python app where you can add a product and it returns a graph where it compares with other products and show optimality over time given prices/power and Tops - Gpu vs Npu. Good hunting…
Seems incompatible with Capitalism. ‘Surveillance-capitalism’ is a huge problem, and no other system in history have gathered so much private data about its citizens. Everything we do on the internet, or in some closed ecosystem, is indirectly monitored and the market system knows all about us, our activities, our social groupings, opinions etc and this knowledge is not used for our benefit. Further, as long as we might influence the system, or as long as we might have digital data ‘belonging’ to someone else, capitalism will step in and take control of our ‘privacy’. We and privacy are a threat to that system…
Paywalled, but interesting topic. Afair, being able to generate internal vision in the same way as our inner voice, is called “Phantasia”. “Aphantasia” is the lack of that ability. I learned ~6 months ago that everybody else could actually close their eyes and see a thing in front of them. However, somehow I can still sense movement and some vague 3D shapes of stuff, and that seem to be enough for me to ‘navigate’ most information.
OC I felt cheated out of ~50 years of opportunities to conjure unwanted images up in other peoples heads ! Cock, sweat, anal, tits, puss, a hot turd between toes, and a million other image generating references that I could have casually snuck into conversations in the most in-appropriate situations … what a waste of a great supervillain power ! “The Nasty-Visions-Man” was not to arise :-(
Anyway, I’m curious to how blind-born people imagine or sense the physical world. They must have their proprioception and a 3d world model, so they should be able to sense movement/shapes etc like me, and imagine how others move. We all have a dozen+ internal representations of the outside world, and besides the potential for human mis-communication etc, it also hints at how to build a full artificial brain, AGI.
Neat. I had to think about the purpose of running a kernel in a browser, and couldn’t find any at first, but if we just see wasm as a browserless container environment, then it makes more sense than running a linux kernel in a browser. I could see a minimal linux spinning up specialized kernels in different wasm containers, much like vm’s. Not sure what hardware/software roadblocks exist in a modern wasm container environment tho.
This will not only change the global power balance away from the insane Empire, it will also change global transport as we know it. Ships, flight or just about anything big that moves or need power will benefit from this. Local mini nuclear power stations may be a thing. That means that Russia have a golden egg here - not quite like AGI, but close.
The Western ‘free market’ religion is getting clubbered - and it does some of the good work it-self ! …and that’s after the West’s Plutocracy have spent/wasted trillions of stolen resources over the last 75 years trying to keep all other descent social systems/nations down. Monetary and political colonization have taken over as the Western Plutocracy’s favorite conquering tool, after several hundred years of using military as a tool for colonization. The amount of information control in nations under the ‘free market’ occupation are staggering, and the price and suffering for the whole world are immeasurable and will always remain as an embarrassing moment in the history of our “Civilization”.
Welcome to a new global multi-polar leadership that will propel us all past the feeble, inefficient and Psychopathic Western society/belief system and culture (my ex-culture…).
…unless, oc, the Psychopathic elite in the West decides to set the world on fire - killing even more millions - trying to rescue their “Empire of Lies”. It won’t take long before the Fascist empire will implode…
It won’t be nice to hear, but here goes… Because the ‘woke’ movement pushed by the Dems, were really just a psyop - propaganda - to create hatred against orthodox Russia/Putin. Remember pussy-riot, the Gay Putin memes etc ? They were all US supported/instigated narratives pushed in Nato + especially Ukraine for many years. Now that the campaign didn’t amount to the expected result, the US plutocracy/MIC don’t need/want it anymore, and not all Dems were onboard with the ‘hippie’ psyop that went against their core morale. They are likely pushing back now.
Obama/Biden and other top Dems knew all along that it were just a campaign to pump up hate in a move against their perceived enemies, and don’t care shit about ordinary people and their sexuality - or life really. I mean, when did US top politicians suddenly grow a desire to defend the little guy ? I’ve never heard of such a thing.
Re. social media/censorship. US spends a ton of resources grooming social media platforms for ‘proper’ western/liberal content/values, and the TikTok issue - among many - are an attempt to control information on that platform also. Specially young people have via uncontrolled media, seen what happens in israel/gaza, Ukraine/Russia, and even seen how China is really like, and even become friends with them. The young people on uncontrolled media (non US sphere of interest) are - according to the US leadership - getting behind the ‘wrong’ party in masses. That made the US Plutocracy flip out and make all kinds of problems for the Chinese company (hundreds of fake rumors, tariffs/pressure, and now money to buy it. The whole shebang were rolled out for ‘National Security’, which oc is just the security of the Plutocracy - not the people…
I think the ‘normal’ prosecuting morale is coming back - unfortunately. …and fascism - that were already here before Drump - will become much worse as the US plutocratic regime decline into recession/crash. Be careful over there !
Wasn’t this the same natural glue the Chinese used for their latest bone2lock (?) bone cement ?
Somehow, I read the headline as if the new glue fractured its own bonds - as a fail, so just adding the sub-headline for clarity. Nifty discovery btw:
“New bio-glue mimics oysters, fixing shattered bones in 3 minutes without major surgery.”
Agree, this is almost too dumb/obvious - even for the 3-letter psychos and the fanatics they play with…