Musk is arrogant to think he, or anyone else for that matter, can speak authoritatively on AI matters. He has done zero (scientific) research in the area and tries to see so far into the future that it's ridiculous. The same goes for Zuckerberg but at least he isn't claiming to be more right than anyone else.
It seems like Rodney Brooks was saying the same thing to Musk last week. Musk is out of his sandbox here. The latest iteration of AI gives us a nice categorization and pattern matching tool which has certain ethical consequences, but when I hear Musk talk about it it seems like he thinks that hard AI is just around the corner. I haven't seen anything to suggest that it is. We don't even have AI as smart as a housefly.
It's not really relevant whether hard AI is or is not just around the corner if it's able to self-improve and set its own goals at a pace we can't follow or understand. The probability of ir happening this afternoon is close to zero, but the probability of it having happened before the end of the century is very close to 1. This means it's possible there will be no meatware celebrating the turn of the century.
It does matter because we have no idea what the world will look like when we do get to a point where we know it's 10-20 years away. Right now we're not even close to there yet and so it's mostly fear mongering.
No one has the right to presume that the future will be similar to the past, neither Elon Musk, Mark Zuckerberg or Nostradamus.
The definition of AI in public discussion has become so loose that it's basically a religious war. AI can mean anything from, "deep learning enabled by neural networks on graphical processing cards," to "software tools that seem intelligent in some way," to literal Lt. Commander Data type technology.
What does anyone mean when they accuse anyone of having a dim understanding of a poorly defined concept? To may they are basically saying, "look at me, pay attention to me and my companies."
I'm not sure why people continue to persist in the idea that AI will magically leap to human and super-human levels of intellect when we have a hard time engineering them to see patterns that it seems the average person can recognize with little training. I'm not suggesting that no such AI can exist but I'm suggesting that we're not at the point of comprehending the components that goes into such an AI (because I think we'll never directly create it but rather build its foundations which then will converge on their own accord and probably by accident).
In any case, I think the key legal problems of current AI and machine learning by extension are related to privacy. We've let such services invade various parts of our lives and we have no legal barriers sufficient to ensure no one's civil rights are violated as a result. So, let's work on those issues before we worry about making the next Colossus or Ultron.
I think the worry is that at some point someone will crack a self-improving general AI and at that point the time between "we did it" and "oh shit" could be on a human timescale very very short.
Given the potential downsides are possibly extreme (however unlikely) I'm happy that people are paying attention.
I think a lot of "AI" is hype and hyperbole and largely those people don't worry me that much, its the quiet ones doing fundamental research that worry me a bit more.
You are talking about a breakthrough with a threat level in the same magnitude as nuclear weapons, dinosaur killing asteroids etc.
Personally if I had to put my money down I'd say I wouldn't see a human-level general intelligence in my life time but a self-improving/learning intelligence doesn't have to be human level to be very very dangerous either.
That's logical, no? While we do not know when that will happen, once it does, it will snowball massively from there on out. It may happen in 5 years, or in 50, or in 150 -- nobody knows. For all we know we could be one seminal paper away from a fundamental tech that will enable GAI.
Due to the democratization of machine learning tools there is insufficient demarcation between the science and cargo scientist culture warriors right now.
In short, it has become political. Suggesting that we should come up with more clear definitions and timelines, or alternately not discuss AI in a way that either wows or scares people is accepted as being "anti-technology," somehow. This is motivated by the high expectations that investors in this space have placed in machine learning tools, and the fear of the overall economy being stuck in proof of concept hell, resulting in a loss of investment.
It's essentially a very sophisticated boiler room.
19 comments
[ 3.2 ms ] story [ 63.4 ms ] threadhttps://openai.com/about/
Just trying to open the floor to discussion, ground rules, and humanism before its too late.
Zuckerburg's view of AI extends into the next fiscal quarter, Musk's extends to other planets.
The definition of AI in public discussion has become so loose that it's basically a religious war. AI can mean anything from, "deep learning enabled by neural networks on graphical processing cards," to "software tools that seem intelligent in some way," to literal Lt. Commander Data type technology.
What does anyone mean when they accuse anyone of having a dim understanding of a poorly defined concept? To may they are basically saying, "look at me, pay attention to me and my companies."
In any case, I think the key legal problems of current AI and machine learning by extension are related to privacy. We've let such services invade various parts of our lives and we have no legal barriers sufficient to ensure no one's civil rights are violated as a result. So, let's work on those issues before we worry about making the next Colossus or Ultron.
Given the potential downsides are possibly extreme (however unlikely) I'm happy that people are paying attention.
I think a lot of "AI" is hype and hyperbole and largely those people don't worry me that much, its the quiet ones doing fundamental research that worry me a bit more.
You are talking about a breakthrough with a threat level in the same magnitude as nuclear weapons, dinosaur killing asteroids etc.
Personally if I had to put my money down I'd say I wouldn't see a human-level general intelligence in my life time but a self-improving/learning intelligence doesn't have to be human level to be very very dangerous either.
In short, it has become political. Suggesting that we should come up with more clear definitions and timelines, or alternately not discuss AI in a way that either wows or scares people is accepted as being "anti-technology," somehow. This is motivated by the high expectations that investors in this space have placed in machine learning tools, and the fear of the overall economy being stuck in proof of concept hell, resulting in a loss of investment.
It's essentially a very sophisticated boiler room.
Recommended read:
https://en.wikipedia.org/wiki/Superintelligence:_Paths,_Dang...
Of course, he's the god king.
Zuck is just some idiot kid who has been building impressive products using machine learning since he was in high school.