3 ms·

How can you speak with such confidence, when Hinton and Ilya Sutskever can not? Hinton even said it is possible in next 5-10 years. While, you can believe what

How can you speak with such confidence, when Hinton and Ilya Sutskever can not? Hinton even said it is possible in next 5-10 years. While, you can believe whatever you believe, please don't bash people as laypeople or something as nonsense.

Manjuuu 4y ago
The same confidence of those already planning for the advent, that will never be, of AGI. I see a lot of unmotivated enthusiasm for the "new" thing, the opinions of Hinton should be considered valid if the reasoning makes sense, not just because the source is Hinton. No idea why he said that it's possible in 10 years. Not trying to bash anyone, I just meant normal people, outside of the field. Enthusiastic nonsense is still nonsense.
- 93po 4y ago
  To say we won't have AGI in our lifetime as a certainty means that you must be able to say with a certainty how AGI is developed. Otherwise there is no evidence to point to as to why the steps towards developing it aren't possible.
- Workaccount2 4y ago
  We have these guys calling themselves "experts on intelligence" just because they know the structure of the components of the neural net. It's like neurologists saying they are experts in consciousness because they know how neurons function and grow. Thankfully doctors don't have nearly the same levels of hubris as tech bros.
  - Manjuuu 4y ago
    Exactly.
- YeGoblynQueenne 4y ago
  >> No idea why he said that it's possible in 10 years. It's because he's been saying that for 40 years.
  - karmasimida 4y ago
    https://youtu.be/qpoRO378qRY https://youtu.be/qpoRO378qRY Hinton said he didn’t believe it before ChatGPT came out, that AGI is possible in 20 years. It is simple to just checkout peoples’ own words, otherwise you are hallucinating the same way as those LLMs do
    - YeGoblynQueenne 4y ago
      Can you please link me to the point in the interview where Hinton says what you say? It's 42 minutes long and I can't know which part of it you're referring to.
YeGoblynQueenne 4y ago
>> How can you speak with such confidence, when Hinton and Ilya Sutskever can not? Hinton even said it is possible in next 5-10 years. Is all the discussion in this thread about the abilities of ChatGPT in 5-10 years, or is it about its abilities right now?
ActorNightly 4y ago
Because nobody even understands what AGI is. Or more charitably, has their own version of what AGI means in their head and nobody realizes it that they are talking about different things. Will we get very good software in the form of personal assistants that could answer questions in depth about a large number of topics? Most likely. Chat GPT is very good at a lot of things, and its only going to get better. Compute amount is going to be an issue though. Will we get something that resembles a human sitting behind a keyboard, except an expert in every single subject on the internet with some markov chain process that determines objective subfunctions to minimize? Could be. Thing is though, there still has to be an overarching code to run that markov chain, which is not AGI. Definitely dangerous in the wrong hands, but not dangerous by itself - CTRL C could stop its execution at any time. Will we get a AGI that is smart enough to circumvent that ctrl c signal, break containment, take over the world, and kill all humans, all in the spirit of turning the earth into a super computer? Nah. To even start to code that AGI, you need to get around the whole principle of computational irreducability which means first proving that P=NP. An AGI will essentially have to run simulations on the world internally to figure out how things could play out, at an information amount threshold that is less than the information in the world its trying to simulate, which (unless P=NP) is impossible.

Manjuuu 4y ago

The same confidence of those already planning for the advent, that will never be, of AGI. I see a lot of unmotivated enthusiasm for the "new" thing, the opinions of Hinton should be considered valid if the reasoning makes sense, not just because the source is Hinton. No idea why he said that it's possible in 10 years. Not trying to bash anyone, I just meant normal people, outside of the field. Enthusiastic nonsense is still nonsense.

93po 4y ago
To say we won't have AGI in our lifetime as a certainty means that you must be able to say with a certainty how AGI is developed. Otherwise there is no evidence to point to as to why the steps towards developing it aren't possible.
Workaccount2 4y ago
We have these guys calling themselves "experts on intelligence" just because they know the structure of the components of the neural net. It's like neurologists saying they are experts in consciousness because they know how neurons function and grow. Thankfully doctors don't have nearly the same levels of hubris as tech bros.
- Manjuuu 4y ago
  Exactly.
YeGoblynQueenne 4y ago
>> No idea why he said that it's possible in 10 years. It's because he's been saying that for 40 years.
- karmasimida 4y ago
  https://youtu.be/qpoRO378qRY https://youtu.be/qpoRO378qRY Hinton said he didn’t believe it before ChatGPT came out, that AGI is possible in 20 years. It is simple to just checkout peoples’ own words, otherwise you are hallucinating the same way as those LLMs do
  - YeGoblynQueenne 4y ago
    Can you please link me to the point in the interview where Hinton says what you say? It's 42 minutes long and I can't know which part of it you're referring to.

YeGoblynQueenne 4y ago

>> How can you speak with such confidence, when Hinton and Ilya Sutskever can not? Hinton even said it is possible in next 5-10 years. Is all the discussion in this thread about the abilities of ChatGPT in 5-10 years, or is it about its abilities right now?

ActorNightly 4y ago

Because nobody even understands what AGI is. Or more charitably, has their own version of what AGI means in their head and nobody realizes it that they are talking about different things. Will we get very good software in the form of personal assistants that could answer questions in depth about a large number of topics? Most likely. Chat GPT is very good at a lot of things, and its only going to get better. Compute amount is going to be an issue though. Will we get something that resembles a human sitting behind a keyboard, except an expert in every single subject on the internet with some markov chain process that determines objective subfunctions to minimize? Could be. Thing is though, there still has to be an overarching code to run that markov chain, which is not AGI. Definitely dangerous in the wrong hands, but not dangerous by itself - CTRL C could stop its execution at any time. Will we get a AGI that is smart enough to circumvent that ctrl c signal, break containment, take over the world, and kill all humans, all in the spirit of turning the earth into a super computer? Nah. To even start to code that AGI, you need to get around the whole principle of computational irreducability which means first proving that P=NP. An AGI will essentially have to run simulations on the world internally to figure out how things could play out, at an information amount threshold that is less than the information in the world its trying to simulate, which (unless P=NP) is impossible.