Am I reading this right? Is this the Microsoft playbook of selling licenses for education? If so brillant move by OpenAI and Anthropic.
You can call it a subculture or philosophy or movement, but social movement to me implies something with broad appeal.
I do not think you can call it a social movement because it does not have any significant mass of people behind it and by its nature, never will.
It's a reflection of the demand from people who want to live in Berlin.
Used to be in every AI/ML job advert, that they preferred those who had previously published in the top 3. Ironically now it has become a bit of a meme and I see it less often as one of the "nice to have".
Even with AI you'd want the AI to be writing python scripts instead of following an analysis.md.
Lots of EU people are now travelling longer distances due to housing prices and job relocations. My last commute in Germany was 1.5 hr one way and some of my colleagues had 2.5 hr commutes. Hybrid hours though.
True although a big component of the energy prices in Europe are due to policy driven taxation. There is a huge difference in intra EU petrol prices from around 6$/gal to 10$/gal.
I like to say, traders and people you put money on the line are often far more honest about economic policy, than economic influencers like The Economist.
Nobody ever recognizes themselves as extremist.
I completely agree with this, but my banking apps, my broker, my health insurance, my simcard provider all already require my face for identification.
What I do not find plausible is this scenario where AI is so smart it can replace us all, but at the same time somehow still easily controllable and utterly subservient to the ownership class.
No because public EU compute is adminisered by physicists and although they are very competent in their own domains they are winging it when it comes to AI. It like expecting OakRidge to train LLMs... They already tried…
I don't think using the median lethal dose of toxicity is a good metric to compare pesticides.
Idk, I often felt the opposite. I naturally tend towards working on hard problems but my feeling is that if I chose instead to focus on an easier "thing" I could execute really well and be overall more successful.
Multiple Python based DSLs, just like now with GPU programming.
I would become a patriotic European too, but only after our leaders at the national and EU level stop being transatlanticist sellouts... which won't happen any time soon.
That you need world models to sensibly deploy "thinking" machines in the real world. Else they do stupid shit like drive straight into water. You can bruteforce some semblance of thinking by training on literally all…
This viewpoint never made sense to me, because you have variation within any culture, and because advantageous strategies win out over time, any culture will evolve. Therefore, it never made sense to me to consider…
LeCun is right.
AI would be able to clue you in on the logical fallacy you're committing.
Why not buy a used one?
Programming languages are not just for ergonomics. They are valuable abstractions that help us reason. And they also help LLMs reason in the same manner.
You jest, but isn't this the logical conclusion? A sufficiently smart AGI has no need for humanity, at all.
Having lived in Germany, the strongest cultural conflict I felt was inflexibility of plans. The German way is to plan something very meticulously and the to follow through with the plan no matter what. I am however of…
Am I reading this right? Is this the Microsoft playbook of selling licenses for education? If so brillant move by OpenAI and Anthropic.
You can call it a subculture or philosophy or movement, but social movement to me implies something with broad appeal.
I do not think you can call it a social movement because it does not have any significant mass of people behind it and by its nature, never will.
It's a reflection of the demand from people who want to live in Berlin.
Used to be in every AI/ML job advert, that they preferred those who had previously published in the top 3. Ironically now it has become a bit of a meme and I see it less often as one of the "nice to have".
Even with AI you'd want the AI to be writing python scripts instead of following an analysis.md.
Lots of EU people are now travelling longer distances due to housing prices and job relocations. My last commute in Germany was 1.5 hr one way and some of my colleagues had 2.5 hr commutes. Hybrid hours though.
True although a big component of the energy prices in Europe are due to policy driven taxation. There is a huge difference in intra EU petrol prices from around 6$/gal to 10$/gal.
I like to say, traders and people you put money on the line are often far more honest about economic policy, than economic influencers like The Economist.
Nobody ever recognizes themselves as extremist.
I completely agree with this, but my banking apps, my broker, my health insurance, my simcard provider all already require my face for identification.
What I do not find plausible is this scenario where AI is so smart it can replace us all, but at the same time somehow still easily controllable and utterly subservient to the ownership class.
No because public EU compute is adminisered by physicists and although they are very competent in their own domains they are winging it when it comes to AI. It like expecting OakRidge to train LLMs... They already tried…
I don't think using the median lethal dose of toxicity is a good metric to compare pesticides.
Idk, I often felt the opposite. I naturally tend towards working on hard problems but my feeling is that if I chose instead to focus on an easier "thing" I could execute really well and be overall more successful.
Multiple Python based DSLs, just like now with GPU programming.
I would become a patriotic European too, but only after our leaders at the national and EU level stop being transatlanticist sellouts... which won't happen any time soon.
That you need world models to sensibly deploy "thinking" machines in the real world. Else they do stupid shit like drive straight into water. You can bruteforce some semblance of thinking by training on literally all…
This viewpoint never made sense to me, because you have variation within any culture, and because advantageous strategies win out over time, any culture will evolve. Therefore, it never made sense to me to consider…
LeCun is right.
AI would be able to clue you in on the logical fallacy you're committing.
Why not buy a used one?
Programming languages are not just for ergonomics. They are valuable abstractions that help us reason. And they also help LLMs reason in the same manner.
You jest, but isn't this the logical conclusion? A sufficiently smart AGI has no need for humanity, at all.
Having lived in Germany, the strongest cultural conflict I felt was inflexibility of plans. The German way is to plan something very meticulously and the to follow through with the plan no matter what. I am however of…