2 ms·

> Just because you list up some type of random technology, doesn't make it an good argument. In what way is this random? It's a chronology of the bullshit fro

> Just because you list up some type of random technology, doesn't make it an good argument. In what way is this random? It's a chronology of the bullshit from the past few years. Also: easy to say in hindsight. The hype was there, and many people believed it. The press, in particular, just lapped up the claims the way a Labrador Retriever gobbles up whatever you put in front of it. > We have opened up a potential pandoras box without knowing it. No, we haven't. We have a new type of (I would say, inferential) computation that is truly astonishing because it can convincingly imitate language and, to a certain extent, intentional action. Feel free to read "generate" or whatever instead if you don't like "imitate." It won't wake up and murder you in your bed at night. It won't take over the world. It won't wipe out humanity. It won't fly to Mars and build data centers there. It won't maximize paperclips. It probably won't even take your job away. It has no consciousness. It has no intentions. And if you look behind the curtain, all you'll see are a few very rich, unscrupulous guys who are stoking the debate with fearmongering because it's strategically advantageous for them to do so. Don't let them take you for a fool. Use GenAI to your advantage while tokens still cost a fraction of what they're actually worth, but ignore all that doomsday nonsense.

pixl97 12d ago
Hey, I remember when the Internet was just a bunch of bullshit that was going to go away too. >It has no consciousness. Why does it need it? >It has no intentions. All it needs is a prompt. Instrumental convergence (and humans hooking them up to everything else) takes care of the rest. Really you sounds horrifically ignorant of how any evolutionary AI algorithms work and the problems we see come out of them in much simpler models.
- ahf8Aithaex7Nai 12d ago
  Yeah, yeah, sure, I'm ignorant. In any case, I recognize unfalsifiability when I see it. So it seems there's still some room for even more ignorance beneath my ignorance.
hollerith 12d ago
Just to focus on one part of your comment: does a program that can reliably beat you or I or Magnus Carlsen at chess have an intention to win chess games?
- mirelahmd 12d ago
  its optimizing the best moves that increases win prob thats it
  - hollerith 12d ago
    I'm worried about what happens when an AI or a coalition of AIs starts optimizing reality instead of just optimizing chess boards.
Gregkion 12d ago
I never followed these hypes and it was very clear very fast that this stuff was bullshit. Did I see that I could get rich with bitcoins investing until 100k and selling everything? For sure not. But i already saw basic flaws of crypto when it came out. Hyperloop? I cried out very early on how dangeriuos this is and how stupid and that no one wants to ride in a small tube without proper precautions (i was wrong here, some poeple really use that las vegas tube). But lets be clear about LLM: The first thing which motivated me to really look into this was a random YouTube video of someone talking/writing to GPT-3 and ask it to simulate being a mysql server. And it simulated this flawless. A table creation statement, 3 inserts and a working select statement. That showed me something: The chance that someone trained that 'thing' to be able to simulate a random mysql server was very low. I then tried it and the first thing not happened: I wasn't able to find the flaw. You know when Google IO or Apple Keynotes are happening and you get your hands on any demo and you live in a differen country or speak a different language or you are just not in San Francisco were most demos were maid for, it breaks fast. GPT-3 didn't break fast. I was able to interleave english with german words, no issue. And i wasn't following LLMs before, i was following progress in machine learning and machine learning started to get more and more traction. I can now prompt a tremendes amount of generic things to a neural network and it responds very convincing. I have never seen technology advancing that fast and its quite clear what we have to still build around this new technology. We are not even were we will be and its obviouse. Only when we build all the stuff (ai layer, ai platforms, ai chips, language and skill specific models, ...) and then starting to see any plateu we can stop keeping an eye on it. And while we do that, everything just got a little bit better again and another interseting new thing was discoveed about LLMs, AI, AGI etc. Right now we are in the middle of a reinforcement learning phase which will bring even more. These 'no consciousness' thing already took jobs from people. I was planing to pay an blender artist money on fiverr for a project, since Opus 5.0 it was just able to do this for me without fiverr. I don't think these unscrupulous guys are wrong. I'm following this area for a decade and we might have hit an inflection point: We have the compute, apparently we have the money as a society to build this massive amout of compute, we have also everything digital. It doesn't matter what it is, everything we have is digital. Every training material, every best practice tips and tricks everythign is digital. You can be as dismissive about all of this as you want, but when the economy of a whole planet invest the most money it ever had into one area of research, you should try to keep an eye of where this is going. And these random technologies? Random people invested in these for different reasons for sure, but the investment into AI is a lot broader with a lot more money behind it. The richest companies on the planet with the smartest people on the planet are doing one thing: Investing in AI.

pixl97 12d ago

Hey, I remember when the Internet was just a bunch of bullshit that was going to go away too. >It has no consciousness. Why does it need it? >It has no intentions. All it needs is a prompt. Instrumental convergence (and humans hooking them up to everything else) takes care of the rest. Really you sounds horrifically ignorant of how any evolutionary AI algorithms work and the problems we see come out of them in much simpler models.

ahf8Aithaex7Nai 12d ago
Yeah, yeah, sure, I'm ignorant. In any case, I recognize unfalsifiability when I see it. So it seems there's still some room for even more ignorance beneath my ignorance.

hollerith 12d ago

Just to focus on one part of your comment: does a program that can reliably beat you or I or Magnus Carlsen at chess have an intention to win chess games?

mirelahmd 12d ago
its optimizing the best moves that increases win prob thats it
- hollerith 12d ago
  I'm worried about what happens when an AI or a coalition of AIs starts optimizing reality instead of just optimizing chess boards.

Gregkion 12d ago

I never followed these hypes and it was very clear very fast that this stuff was bullshit. Did I see that I could get rich with bitcoins investing until 100k and selling everything? For sure not. But i already saw basic flaws of crypto when it came out. Hyperloop? I cried out very early on how dangeriuos this is and how stupid and that no one wants to ride in a small tube without proper precautions (i was wrong here, some poeple really use that las vegas tube). But lets be clear about LLM: The first thing which motivated me to really look into this was a random YouTube video of someone talking/writing to GPT-3 and ask it to simulate being a mysql server. And it simulated this flawless. A table creation statement, 3 inserts and a working select statement. That showed me something: The chance that someone trained that 'thing' to be able to simulate a random mysql server was very low. I then tried it and the first thing not happened: I wasn't able to find the flaw. You know when Google IO or Apple Keynotes are happening and you get your hands on any demo and you live in a differen country or speak a different language or you are just not in San Francisco were most demos were maid for, it breaks fast. GPT-3 didn't break fast. I was able to interleave english with german words, no issue. And i wasn't following LLMs before, i was following progress in machine learning and machine learning started to get more and more traction. I can now prompt a tremendes amount of generic things to a neural network and it responds very convincing. I have never seen technology advancing that fast and its quite clear what we have to still build around this new technology. We are not even were we will be and its obviouse. Only when we build all the stuff (ai layer, ai platforms, ai chips, language and skill specific models, ...) and then starting to see any plateu we can stop keeping an eye on it. And while we do that, everything just got a little bit better again and another interseting new thing was discoveed about LLMs, AI, AGI etc. Right now we are in the middle of a reinforcement learning phase which will bring even more. These 'no consciousness' thing already took jobs from people. I was planing to pay an blender artist money on fiverr for a project, since Opus 5.0 it was just able to do this for me without fiverr. I don't think these unscrupulous guys are wrong. I'm following this area for a decade and we might have hit an inflection point: We have the compute, apparently we have the money as a society to build this massive amout of compute, we have also everything digital. It doesn't matter what it is, everything we have is digital. Every training material, every best practice tips and tricks everythign is digital. You can be as dismissive about all of this as you want, but when the economy of a whole planet invest the most money it ever had into one area of research, you should try to keep an eye of where this is going. And these random technologies? Random people invested in these for different reasons for sure, but the investment into AI is a lot broader with a lot more money behind it. The richest companies on the planet with the smartest people on the planet are doing one thing: Investing in AI.