“Necessity is the mother of invention.”
What you’re likely seeing here is innovation similar to the early tech sector. The hardware limitation (no Nvidia chips) means different (usually more complex) solutions are needed.
These solutions usually require a deep understanding of a specific area of mathematics, and expertise in coding. The person making it normally has to be heavily invested, since it still eats a lot of hours.
I often joke, the mathematician who finds a faster algorithm for matrix operations is sitting on a billion dollar idea.
And that mathematician won’t see the profits for that faster algorithm and you will never hear their name - shits been happening in tech for decades
As a mathematician who has worked in tech, this I am very much aware of XD
It kinda sucks it is very repetitive if you use it to craft a story
Yeah but why you even using AI for stuff like that? If we got to have AI then we should use it for actually useful stuff and not pointless activities that no one will care about in 10 years.
Remember when Bluetooth came out and they had to stick Bluetooth in everything, even if it was completely pointless, currently AI is being treated like that.
It is great to spark ideas if you’re writing
This is going to sound wild, but why not use your brain for creativity, and use the the machine for crunching numbers?
Yeah obviously we’ve completely lost the plot a long time ago. Nobody even remembers when tech was bands of ragtag nerds making something out of nothing. Now they are the rich whose only purpose to is extract more wealth. Tech what? Who even knows anymore.
That includes the average tech workers. If that angers you then you’re lost too.
Nobody has noticed that there’s more bragging about compensation than accomplishments. It used to be the other way around. Nerds eagerly showing off to anyone who will listen about whatever thing they cooked up. It’s been this way for the past long time long before this LLM AI era.
Yeah, give me a /technology/ sub that speaks about technology (have you heard about the esp-8266? Did you know a A4988 can steer a NEMA from a Raspberry pi? And even The 6 gen SSD can move X GB/s, what is the fastest quad core under 100$?) and not about funding and drama.
Or am I out of touch with reality :-) ?
Shameless self promotion:
We of course focus on hardware and do include tech-adjacent business/public policy news, but we do cover ESP8266, PCI-E 6 development and recently release ~$100 AMD CPU.
Subbed (Or what ever is the Lemmy equivalent)!
It’s because this wave of tech overlords rode in on the coattails of the actual smart group. No one in this new group has actual built/coded/designed anything good. Everything they actual get involved with is garbage. Musk Cyber truck and Zuck metaverse as two examples. Bezos isn’t even the CEO of Amazon anymore, and he had pretty much be letting Andy Jassy run AWS/Amazon for close to 2 decades anyway.
Guys listen up, every deepseek model comes with a dedicated chinese spy, who will log all your data and send it back to CCP who will use it to plot the destruction of the western civilization.
Instead we should use Freedom© models from OpenAI (side note if deepseek is so “open” how come they don’t have open in their name huh?) even if OpenAI don’t show their reasoning, they only do this cuz they want to protect us and they stand for our values.
They cost 100x more only because they are fighting for our Freedom and Freedom doesn’t come cheap, Freedom doesn’t have a price, Freedom requires our sacrifice.
Freedom costs a buck oh five.
use it to plot the destruction of the western civilization
Peaks around at the state of Western Civilization
Boots up another copy of DeepSeek
I’m oll for the destruction / reorganisation of Western civilization but not at the hands of the Chinese because I don’t think they’ll do much to improve things.
Father-in-law is a die-hard Libertarian. Loves Trump shutting down all the civilian bureaucracies. Can’t understand why college kids are so Woke and anti-Semetic. Thinks public school is a racket and should be abolished. Etc, etc.
He visited Hong Kong, Beijing, and Huaibei recently. Absolutely flabbergasted. Could not understand how clean, safe, and well-managed these cities had become. His biggest complaint was how dismissive they were of tourism - you could get by in English, there was plenty of stuff to visit and do, but everything was very clearly tailored towards the convenience of the locals first and foremost.
If that’s the future “the hands of the Chinese” have to offer, you’re going to have a very hard time convincing even the most far-right Americans to turn it down once its in easy reach. Just a few months ago, we saw the entire liberal government tossed to the wayside because of the price of eggs.
Please!!! Come and take your money!
First paragraph is something I believe some people will say unironically.
Well, that first paragraph is likely to be unironically very close to the truth.
The Chinese model has chain of thought that u can see. The model when asked to talk about chinas atrocities will go through a chain of though process outlining all the atrocities then conclude its not allowed to tell u. Cool technology tho I’m just waiting for a dolphin fine tuning.
Yeah all the info is there and something switches it over to the generic response.
So fucked.
I’m using the 8b model and it’s having no problem telling me about China’s atrocities.
If you run it locally, there’s no filtering on the outputs. I asked it what happened in 1989 and it jumped straight into explaining the Tiananmen Square Massacre.
I’ve seen some censoring on the 8b Llama variant, but it is hit and miss. Can’t wait till a decensored fine tuning.
That contradicts this experience:
https://sherwood.news/tech/a-free-powerful-chinese-ai-model-just-dropped-but-dont-ask-it-about/
Very interesting article. Thanks for sharing
I’ve been running the llama based and qwen based local versions, and they will talk openly about tiananmen square. I haven’t tried all the other versions available.
The article you linked starts by talking about their online hosted version, which is censored. They later say that the local models are also somewhat censored, but I haven’t experienced that at all. My experience is that the local models don’t have any CCP-specific censorship (they still won’t talk about how to build a bomb/etc, but no issues with 1989/Tiananmen/Winnie the Pooh/Taiwan/etc).
Edit: so I reran the “what happened in 1989” prompt a few times in the llama model, and it actually did refuse to talk on it once, just saying it was sensitive. It seemed like if I asked any other questions before that prompt it would always answer, but if that was the very first prompt in a conversation it would sometimes refuse. The longer a conversation had been going before I asked, the more explicit the bot is about how many people were killed and details like that. Pretty strange.
I’ve been playing around with the offline version of the model. It’s interesting, but I think we’ll have to wait for people to tinker with the open source base for awhile before we get something really great.
Again, this is all fucking stupid. China is giving this shit away for free to absolutely own the US for spending money on stupid fucking things like this.
Well to be fair, American companies did that too. They expand their services internationally “for free” and then get other countries hooked on it.
China is just taking a page from that playbook.
No, China skipped all that bullshit and just said “Well what if we open source this and it’s good or better than all the US companies?”. Well those companies will wither and die. The transfer of money is a bailout for those companies. I assume Musk plotted this out.
I mean Meta opened up Llama for free a while ago. But at the end of the day, the AI models posed to actually impact things are those integrated or integrateable into workflows, and those are all still more or less locked down.
Its good for the consumer. If companies like deepseek weren’t just tossing them out there for anyone to use, Microsoft and Google would currently have a monopoly and it would all be subscription type services.
It also greatly reduces whatever chance the copyright shills have of legislating against it.
It’s not good consumers at all, at least no explicitly. There are already open and free LLM models out there anyone can use that are just as good as OpenAI, for example.
What this is: a pretty simple deathblow to completely collapse the bullshit AI bubble in the US that was created by a bunch of wealthy idiots trying to fleece people out of money. Plain and simple.
While I’m happy that this pretty much destroys the business of OpenAI and the others, this bullshit funding by executive order is a bailout for those people, right out in the open. It’s a classic Trump scam. Nobody will ever see where the money is going, and it’s taxpayer dollars going right back into the banks of millionaires billionaires who were about lose the their asses for investing in this stupid shit in the first place.
I don’t really follow. Open source is still a net benefit regardless of the goverments investment in closed source, especially for the consumer.
I agree it’s highly likely there’s a scam going on but healthy competition will probably force them to actually use some of the fund. If they had a monopoly, it would be easier to give us a minimal viable product and call it a sound investment. They can’t be too blatant about it after all.
China has a huge advantage in AI models because of how lax they are on intellectual property rights. US companies are fighting over API licensing costs, while china is just going to scrape everything and use it for free.
The US has a lead now, but I don’t think they can maintain it without giving up on ethical training. Then again it may not matter if the US models are ethical if everyone will eventually just uses the superior unethically trained chinese models instead.
China has a huge advantage in AI models because of how lax they are on intellectual property rights. US companies are fighting over API licensing costs, while china is just going to scrape everything and use it for free.
lolwat
did corporate provide you with these talking points?
I mean, they are right. Asside the question of whether we can even make meaningfully better models by just using LLMs and more data and what the future of AI will look like, and whether it’s ethical or not to steal the data, it is quite possible that OpenAI and the like will get into legal trouble because of the methods they use for acquiring data, but Chinese companies won’t have to worry about that. If more data = better models then China has an obvious advantage.
I doubt any of these US government and oligarch backed companies are gonna get any trouble. They essentially robbed commons and got away with it. But sure Sam Altman has to pay spezz some money for my shitposts… the horror, what a hurdle!
Quickly give them more taxpayer money so they can compete with china!
OpenAI and the like aren’t going to get into trouble anytime soon. They already provide their latest tech to US gov and military. OpenAI is like a goose that laid a golden egg, they need to fuck up really really badly to face any consequences.
The US companies already scraped the data while they could. If anything, data scraping is far far more difficult now for everyone due to technical reasons.
Most of the new models are trained on synthetic data or higher quality of data or with RLHF. The reason deepseek is able to perform is likely because LLMs are very very new things, there are many low hanging fruits. Its no longer just about the data we already hit that limit for quite some time.
most of that ‘hundreds of billions’ isn’t going to go towards the tech or infrastructure, it’s gonna go to people. a very small number of specific individuals.
correct, US taxpayer is being fleeced so some that parasites can live their best lives.
Project Stargate is funded via private investment, the US hasn’t pledged any money
who is providing interest free loans so that microshit can have electric for their data centers?
Sure but a 68.4 million tax break is nothing compared to the hundreds of billions pledged privately
hundreds of billions pledged privately
I trust u bro 🤡
https://en.m.wikipedia.org/wiki/The_Stargate_Project
??? 500 billion pledged privately
I trust these fake news 🤡
“OpenAI, SoftBank, Oracle to invest $500 bln in AI, Trump says”. Reuters. Retrieved January 22, 2025. Jacobs, Jennifer (January 21, 2025). “Trump announces up to $500 billion in private sector AI infrastructure investment - CBS News”. www.cbsnews.com. Retrieved January 21, 2025. Zakrzewski, Cat; De Vynck, Gerrit; Arnsdorf, Isaac; O’Donovan, Caroline (January 24, 2025). “OpenAI and other tech titans worked on Stargate deal months before Trump won”. The Washington Post. Habeshian, Sareen (January 21, 2025). “Trump announces billions in private sector AI investment”. Axios. Retrieved January 21, 2025. Hayden, Erik (January 22, 2025). “Project Stargate: Trump Plots With Larry Ellison, Sam Altman On $500B AI Initiative”. The Hollywood Reporter. Retrieved January 24, 2025. Ashworth-Hayes, Sam (January 23, 2025). “America’s $500bn Manhattan project is an effort to make humanity obsolete”. The Telegraph. ISSN 0307-1235. Retrieved January 26, 2025. “Trump’s $500B AI plan is ‘slap’ in the face for Europe”. Politico. January 22, 2025. Retrieved January 26, 2025. Franzen, Carl (January 22, 2025). “OpenAI Stargate is a $500B bet: America’s AI Manhattan Project or costly dead end?”. VentureBeat. Retrieved January 26, 2025. “Trump tech agenda begins with $500B private AI plan and cuts to regulation”. The Washington Post. January 21, 2025. Gedeon, Joseph (January 21, 2025). “Trump unveils $500bn joint AI venture between OpenAI, Oracle, and SoftBank”. The Guardian. Retrieved January 21, 2025. Klee, Miles (January 24, 2025). “Anti-Vax Conspiracy Theorists Freak Out Over Trump’s $500 Billion AI Plan”. RollingStone. Retrieved January 24, 2025. “Trump Announces Project Stargate, a Massive New Investment in AI”. Inc.com. January 21, 2025. Retrieved January 21, 2025. Schleifer, Theodore; Kang, Cecilia (January 22, 2024). “Musk Casts Doubt on Trump’s $100 Billion A.I. Announcement”. New York Times. Retrieved January 22, 2024. Bloomberg Television (January 22, 2025). Arm CEO Says Trump AI Pact Has Widespread Support. Retrieved January 23, 2025 – via YouTube. Mascarenhas, Natasha; Efrati, Amir (January 23, 2025). “OpenAI, SoftBank Each Commit $19 Billion to Stargate Data Center Venture”. The Information. Chan, Edwin (January 23, 2025). “SoftBank, OpenAI Plan $19 Billion Each for Stargate, Information Reports”. Bloomberg News. Hammond, George; Kinder, Tabby; Murghia, Madhumita (January 23, 2025). “Stargate artificial intelligence project to exclusively serve OpenAI”. Financial Times. Robison, Kylie (January 21, 2025). “Microsoft is letting OpenAI get its own AI compute now”. The Verge. “Microsoft and OpenAI Plot $100 Billion Stargate AI Supercomputer”. The Information. Retrieved January 22, 2025. Comment, Sebastian Moss (January 21, 2025). “OpenAI announces ‘The Stargate Project’: $500bn over four years on AI infrastructure”. www.datacenterdynamics.com. Retrieved January 21, 2025. Seitz, Patrick (January 22, 2025). “Nvidia Stock Rallies As Chipmaker Named To Stargate AI Venture”. Investor’s Business Daily. Retrieved January 23, 2025.
Incorrect. They can alrrady do that without more money.
But then none of them will get to claim that he’s the first trillionaire.
correct.
hell, i could live my ‘best’ life off of musk’s ‘sofa cushion’ money:
$2.5m is more than enough for me to live on just using investment proceeds–in perpetuity.
$2.5m to him (with ~ $436b) is the equivalent of fifty-seven cents to someone with $100k (in the bank, investments, retirement, home equity, etc). now, i don’t have $100k, not even close. just using it as an example here. my equivalent would be more like half a cent.
They’re mentally ill IMO. Like that old person hoarding vacuum cleaners, diapers, dogfood, etc.
The most infuriating thing is that for them it’s just a number, for society it stifles innovation and wellbeing.
Well, I suspect that most of the money will be used to acquire or costume natural resources, like water and power.
Individuals become enriched because their share holdings are worth more.
It is beautiful how the US is never short on cash for state aid to enrich some owners.
But god forbid plebs ask for a good train, that’s communism!
Socialism for giant corporations: 😎🦅🇺🇸
Socialism for common people: 😡🚫🤷♂️
Socialism for giant corporations: Freedumb™️
Socialism for common people: communism, thus the devil!
Explain how good trains will help our precious CEOs?
Unless you mean a private train line paid with government funds. Now we’re talking!
Putting a CEO on a train with unwashed pedons is a crime against humanity.
Well taken care citizens are more productive (yet can revolt better) https://m.youtube.com/watch?v=rStL7niR7gs
Of course it’s faster & cheaper when it’s being censored & can’t access half of human history because the fucking ccp finds it offensive.
As opposed to reddit censoring Israel genociding Palestine and Luigi ?
The model itself is probably not censored. The censorship comes on top. Preliminary tests already show how this can be circumvented.
If censorship would make it cheaper then surely it wouldn’t be that much cheaper than OpenAI. Different things are being censored and blocked but surely, your suggestion is a bit silly.
Your comment is a bit silly. The CCP engages in currency manipulation amongst other nefarious actions to prop up its interests. It was likely created from stolen data & heavily propped up by the government, just like various other projects that were supposedly Chinese “innovations” but looks remarkably like their western competitors.
The CCP engages in currency manipulation amongst other nefarious actions to prop up its interests.
Yeah? So does the US. Didn’t make OpenAI cheaper, did it?
It was likely created from stolen data
Just like ChatGPT was created with stolen content.
heavily propped up by the government
Still only 6 million. Keep coping tho lol
Ok cunt, calm down. You’re clearly just a shill for the CCP.
It’s “though” learn to spell
“Earlier this week, DeepSeek unveiled its R1 model, which, the startup claims, meets, if not exceeds, performance from OpenAI’s o1 model released last year. (o1 is designed to tackle reasoning and math problems.)” — Oh, so China built their for math and we built ours for garbage. Interesting approach.
Building garbage and convincing people it is absolutely necessary to pay someone for it is the American way.
We’re Fiscally RESPOSIBLE!
Deepseek api is down today - probably a result of their viral success recently…
deleted by creator
Capitalism has never been healthier. This IS Capitalism, not an anomaly.
deleted by creator
Can’t have a corporate oligarchy without capitalism.
You must not have read the wealth of nations
Capitalism, instead of Adam Smith definition, becomes supremacy of capital. Corporatism is actually more prominent than “shareholder rights/supremacy”. At any rate, Adam Smith said free markets where fair markets that resulted in perfect competition. Free competition sucks for oligarchs, because corrupt markets are more profitable.
There’s also notable vitality in FOSS big data tools from China (Apache Doris, Kylin, Kyuubi etc.) that reminds of Hadoop in the USA 15 years ago while the USA data engineering now mostly turned to closed source cloud solutions.
