3 ms·

I have little fear of unfriendly AI for two reasons. One is that there is no reason to think artificial intelligence would work differently than natural variety

I have little fear of unfriendly AI for two reasons. One is that there is no reason to think artificial intelligence would work differently than natural variety and therefore could be be policed in similar ways: limiting access and physical opportunity to cause trouble. Plus keeping a watchful eye and deactivating anything worrisome. The other is that the fear of unfriendly AI is being taken unaware by a sudden implementation of hyper-intelligent AIs. So far we haven't made anything smarter than a crab so I doubt we are in imminent danger. There may be a lingering belief that hyper-intelligence could be acquired in a quantum leap forward - skipping past all intermediate levels of intelligence. If we have learned anything from the eternal AI winter it is that advancements in AI are done via painstakingly small improvements.

mhb 12y ago
The AI-Box Experiment: http://yudkowsky.net/singularity/aibox/ http://yudkowsky.net/singularity/aibox/
eli_gottlieb 12y ago
>One is that there is no reason to think artificial intelligence would work differently than natural variety Of course it will work differently from the natural variety. You have dozens of different modules cooperating in your brain to form your mind. It has only a small handful. Your mind runs on heuristics and biases, rarely employing its full power for energy-consumption reasons. An AI just runs at full brainpower all the time and pays for electricity. Go read about how reinforcement learners actually work.
- protonfish 12y ago
  Yes I am aware of all of this and I disagree. Maybe we need to admit that our failure to develop any decent AI is a direct consequence of flaws in our current thinking. Or maybe you should continue to make the same mistakes and expect a different result.
  - eli_gottlieb 12y ago
    Who says that we've terminally failed to develop "any decent AI"? This once again indicates that you don't keep up with the field.
    - IanCal 12y ago
      Usually, "decent AI" is defined as "stuff that humans can do that computers can't currently do". This list is updated as and when computers become capable of doing something, sometimes just to add "Computers can't do X the way a human does X".

mhb 12y ago

The AI-Box Experiment: http://yudkowsky.net/singularity/aibox/ http://yudkowsky.net/singularity/aibox/

eli_gottlieb 12y ago

>One is that there is no reason to think artificial intelligence would work differently than natural variety Of course it will work differently from the natural variety. You have dozens of different modules cooperating in your brain to form your mind. It has only a small handful. Your mind runs on heuristics and biases, rarely employing its full power for energy-consumption reasons. An AI just runs at full brainpower all the time and pays for electricity. Go read about how reinforcement learners actually work.

protonfish 12y ago
Yes I am aware of all of this and I disagree. Maybe we need to admit that our failure to develop any decent AI is a direct consequence of flaws in our current thinking. Or maybe you should continue to make the same mistakes and expect a different result.
- eli_gottlieb 12y ago
  Who says that we've terminally failed to develop "any decent AI"? This once again indicates that you don't keep up with the field.
  - IanCal 12y ago
    Usually, "decent AI" is defined as "stuff that humans can do that computers can't currently do". This list is updated as and when computers become capable of doing something, sometimes just to add "Computers can't do X the way a human does X".