When that major drama unfolded with him getting booted then re-hired. It was super fucking obvious that it was all about the money, the data, and the salesmanship
He is nothing but a fucking tech-bro. Part Theranos, part Musk, part SBF, part (whatever that pharma asshat was), and all fucking douchebag.
AI is fucking snake oil and an excuse to scrape every bit of data like it’s collecting every skin cell dropping off of you.
I’d agree the first part but to say all Ai is snake oil is just untrue and out of touch.
There are a lot of companies that throw “Ai” on literally anything and I can see how that is snake oil.
But real innovative Ai, everything to protein folding to robotics is here to stay, good or bad. It’s already too valuable for governments to ignore. And Ai is improving at a rate that I think most are underestimating (faster than Moore’s law).
I think part of the difficulty with these discussions is that people mean all sorts of different things by “AI”. Much of the current usage is that AI = LLMs, which changes the debate quite a lot
No doubt LLMs are not the end all be all. That said especially after seeing what the next gen ‘thinking models’ can do like o1 from ClosedAI OpenAI, even LLMs are going to get absurdly good. And they are getting faster and cheaper at a rate faster than my best optimistic guess 2 years ago; hell, even 6 months ago.
Even if all progress stopped tomorrow on the software side the benefits from purpose built silicon for them would make them even cheaper and faster. And that purpose built hardware is coming very soon.
Open models are about 4-6 months behind in quality but probably a lot closer (if not ahead) for small ~7b models that can be run on low/med end consumer hardware locally.
I don’t doubt they’ll get faster. What I wonder is whether they’ll ever stop being so inaccurate. I feel like that’s a structural feature of the model.
May I ask how you’ve used LLMs so far? Because I hear that type of complaint from a lot of people who have tried to use them mainly to get answers to things, or maybe more broadly to replace their search engine, which is not what they’re best suited for, in my opinion.
Martin Shkreli is the scumbag’s name you’re looking for.
From wikipedia: He was convicted of financial crimes for which he was sentenced to seven years in federal prison, being released on parole after roughly six and a half years in 2022, and was fined over 70 million dollars
When that major drama unfolded with him getting booted then re-hired. It was super fucking obvious that it was all about the money, the data, and the salesmanship He is nothing but a fucking tech-bro. Part Theranos, part Musk, part SBF, part (whatever that pharma asshat was), and all fucking douchebag.
AI is fucking snake oil and an excuse to scrape every bit of data like it’s collecting every skin cell dropping off of you.
I’d agree the first part but to say all Ai is snake oil is just untrue and out of touch. There are a lot of companies that throw “Ai” on literally anything and I can see how that is snake oil.
But real innovative Ai, everything to protein folding to robotics is here to stay, good or bad. It’s already too valuable for governments to ignore. And Ai is improving at a rate that I think most are underestimating (faster than Moore’s law).
I think part of the difficulty with these discussions is that people mean all sorts of different things by “AI”. Much of the current usage is that AI = LLMs, which changes the debate quite a lot
No doubt LLMs are not the end all be all. That said especially after seeing what the next gen ‘thinking models’ can do like o1 from
ClosedAIOpenAI, even LLMs are going to get absurdly good. And they are getting faster and cheaper at a rate faster than my best optimistic guess 2 years ago; hell, even 6 months ago.Even if all progress stopped tomorrow on the software side the benefits from purpose built silicon for them would make them even cheaper and faster. And that purpose built hardware is coming very soon.
Open models are about 4-6 months behind in quality but probably a lot closer (if not ahead) for small ~7b models that can be run on low/med end consumer hardware locally.
I don’t doubt they’ll get faster. What I wonder is whether they’ll ever stop being so inaccurate. I feel like that’s a structural feature of the model.
May I ask how you’ve used LLMs so far? Because I hear that type of complaint from a lot of people who have tried to use them mainly to get answers to things, or maybe more broadly to replace their search engine, which is not what they’re best suited for, in my opinion.
What are they best suited for?
Martin Shkreli is the scumbag’s name you’re looking for.
From wikipedia: He was convicted of financial crimes for which he was sentenced to seven years in federal prison, being released on parole after roughly six and a half years in 2022, and was fined over 70 million dollars
Thank you!
It’s not snake oil. It is a way to brute force some problems which it wasn’t possible to brute force before.
And also it’s very useful for mass surveillance and war.