Cheaper AI isn’t the pop we want. We want companies to stop trying to use AI for every god damn thing it is terrible at. We don’t want cheaper AI that’s just going to be baked into more stuff.
I mean, I do want cheaper GenAI in the sense that I want people to see that it’s dollar store crap that’s not worth the electricity to run the servers to make it and give it up like they did the fucking Juicero and every other smart appliance a couple years ago. God forbid I hold my breath and people wise up and understand that these people are all grifters looking to tape a horn to a horse and sell their “unicorn” to FAANG or whatever the equivalent is these days, I can’t be assed to rewrite the new poob acronym.
AI isn’t the enemy, though, and we aren’t the ones being grifted—that would be companies who think they can make tons of money replacing people with AI. It’s a reasonable useful tool/fun toy/interesting curiosity in certain circumstances. And for an end user it doesn’t use any more power than a video game. But it’s a tool for craftsmen and folks who understand the limitations, not a replacement for workers. And it sure as hell isn’t a production feature. Anyone looking to make money baking AI into consumer products is an idiot and going to lose their hat.
Nah, o1 has been out how long? They are already on o3 in the office.
It’s completely normal a year later for someone to copy their work and publish it.
It probably cost them less because they probably just distilled o1 XD. Or might have gotten insider knowledge (but honestly how hard could CoT fine tuning possibly be?)
Deepseek showed that actually putting thought into the architecture achieves much more than just throwing more hardware at the problem.
This means a) there will be much less demand for hardware, since much more could be run locally on regular consumer devices. And b) the export restrictions don’t really work and instead force China to create actually better models.
That means, a lot of the investments into the thousands of AI companies are in jeopardy.
I think “just writing better code” is a lot harder than you think. You actually have to do research first you know? Our universities and companies do research too. But I guarantee using R1 techniques on more compute would follow the scaling law too. It’s not either or.
Realistically, the CCP is probably throwing a lot of money at developers to get something good going and available, and US companies are whining about how it’s not fair. The fact of the matter is that a solid product is available for much cheaper, and US companies are now screaming foul. Guess what, a superior product made of good code (people) beats out just throwing money at hardware, who’d’ve gone an thunk it.
What I find really fascinating here is that obviously openAI, Meta, etc. seem to be structurally incapable of actually innovating at this point.
I mean, reducing training costs by literally an order of magnitude just by writing better software is astonishing and shows how complacent the large corporations have gotten.
Meta? The one that released Llama 3.3? The one that actually publishes its work? What are you talking about?
Why is it so hard to believe that deepseek is just yet another amazing paper in a long line of research done by everyone. Just because it’s Chinese? Everyone will adapt to this amazing innovation and then stagnate and throw compute at it until the next one. That’s how research works.
Not to mention China has billions of more people to establish a research community…
ahh… the sound of a bursting bubble.
- plop -
Cheaper AI isn’t the pop we want. We want companies to stop trying to use AI for every god damn thing it is terrible at. We don’t want cheaper AI that’s just going to be baked into more stuff.
I mean, I do want cheaper GenAI in the sense that I want people to see that it’s dollar store crap that’s not worth the electricity to run the servers to make it and give it up like they did the fucking Juicero and every other smart appliance a couple years ago. God forbid I hold my breath and people wise up and understand that these people are all grifters looking to tape a horn to a horse and sell their “unicorn” to FAANG or whatever the equivalent is these days, I can’t be assed to rewrite the new poob acronym.
AI isn’t the enemy, though, and we aren’t the ones being grifted—that would be companies who think they can make tons of money replacing people with AI. It’s a reasonable useful tool/fun toy/interesting curiosity in certain circumstances. And for an end user it doesn’t use any more power than a video game. But it’s a tool for craftsmen and folks who understand the limitations, not a replacement for workers. And it sure as hell isn’t a production feature. Anyone looking to make money baking AI into consumer products is an idiot and going to lose their hat.
Nah, o1 has been out how long? They are already on o3 in the office.
It’s completely normal a year later for someone to copy their work and publish it.
It probably cost them less because they probably just distilled o1 XD. Or might have gotten insider knowledge (but honestly how hard could CoT fine tuning possibly be?)
Deepseek showed that actually putting thought into the architecture achieves much more than just throwing more hardware at the problem.
This means a) there will be much less demand for hardware, since much more could be run locally on regular consumer devices. And b) the export restrictions don’t really work and instead force China to create actually better models.
That means, a lot of the investments into the thousands of AI companies are in jeopardy.
I think “just writing better code” is a lot harder than you think. You actually have to do research first you know? Our universities and companies do research too. But I guarantee using R1 techniques on more compute would follow the scaling law too. It’s not either or.
Realistically, the CCP is probably throwing a lot of money at developers to get something good going and available, and US companies are whining about how it’s not fair. The fact of the matter is that a solid product is available for much cheaper, and US companies are now screaming foul. Guess what, a superior product made of good code (people) beats out just throwing money at hardware, who’d’ve gone an thunk it.
What I find really fascinating here is that obviously openAI, Meta, etc. seem to be structurally incapable of actually innovating at this point.
I mean, reducing training costs by literally an order of magnitude just by writing better software is astonishing and shows how complacent the large corporations have gotten.
Meta? The one that released Llama 3.3? The one that actually publishes its work? What are you talking about?
Why is it so hard to believe that deepseek is just yet another amazing paper in a long line of research done by everyone. Just because it’s Chinese? Everyone will adapt to this amazing innovation and then stagnate and throw compute at it until the next one. That’s how research works.
Not to mention China has billions of more people to establish a research community…
You can write off hardware purchases, paying for skilled devs is like pulling teeth