Posts: 7392
Threads: 53
Joined: January 15, 2015
Reputation:
88
RE: The future of AI?
March 27, 2016 at 4:23 pm
(March 27, 2016 at 4:09 pm)Alex K Wrote: To hardcode these restrictions on the AI into a ROM the problem is that you'd need to know how exactly, in the artificial neural network, the concepts of Humans, injury etc are represented. That's not at all obvious in a deep learning system and may be impossible to determine uniquely let alone beforehand.
Nicely explained.
The very first thing that AI researchers did was to try using logic to reason about the world and interact with it.
This is why we call it classical A.I. or Good Old Fashioned AI (GOFAI).
The problem with using logic is that you need to try dividing up a noisy, messy world into true or false.
Posts: 4196
Threads: 60
Joined: September 8, 2011
Reputation:
30
RE: The future of AI?
March 27, 2016 at 5:35 pm
Like I said, it runs on the interrupt system and the ROM program would check and deal with the scenario if necessary. It obviously is not going to be a 4K ROM. It would not be any harder, perhaps easier, than programming a 'smart car' not to run over people. It will be a while before there are robots that will actually need these laws, but now is the time to start building and debugging the algorithm.
You make people miserable and there's nothing they can do about it, just like god.
-- Homer Simpson
God has no place within these walls, just as facts have no place within organized religion.
-- Superintendent Chalmers
Science is like a blabbermouth who ruins a movie by telling you how it ends. There are some things we don't want to know. Important things.
-- Ned Flanders
Once something's been approved by the government, it's no longer immoral.
-- The Rev Lovejoy
Posts: 18510
Threads: 129
Joined: January 19, 2014
Reputation:
90
RE: The future of AI?
March 28, 2016 at 12:29 am
(March 27, 2016 at 5:35 pm)IATIA Wrote: Like I said, it runs on the interrupt system and the ROM program would check and deal with the scenario if necessary. It obviously is not going to be a 4K ROM. It would not be any harder, perhaps easier, than programming a 'smart car' not to run over people. It will be a while before there are robots that will actually need these laws, but now is the time to start building and debugging the algorithm.
You'll have to be more specific than that it runs "on an interrupt system". How does the system notice that the AI intends to hurt humans?
The fool hath said in his heart, There is a God. They are corrupt, they have done abominable works, there is none that doeth good.
Psalm 14, KJV revised edition
Posts: 4196
Threads: 60
Joined: September 8, 2011
Reputation:
30
RE: The future of AI?
March 28, 2016 at 12:57 am
(March 28, 2016 at 12:29 am)Alex K Wrote: (March 27, 2016 at 5:35 pm)IATIA Wrote: Like I said, it runs on the interrupt system and the ROM program would check and deal with the scenario if necessary. It obviously is not going to be a 4K ROM. It would not be any harder, perhaps easier, than programming a 'smart car' not to run over people. It will be a while before there are robots that will actually need these laws, but now is the time to start building and debugging the algorithm.
You'll have to be more specific than that it runs "on an interrupt system". How does the system notice that the AI intends to hurt humans?
Microprocessors have an interrupt input that will force it to jump to a subroutine. This subroutine would check for interaction with a human. It would be as simple as having a clock circuit that would send an interrupt every 100ms or whatever is deemed appropriate. I take it that you are not involved with microprocessors at the machine level. The 'official' name for such a circuit is " Watchdog timer".
You make people miserable and there's nothing they can do about it, just like god.
-- Homer Simpson
God has no place within these walls, just as facts have no place within organized religion.
-- Superintendent Chalmers
Science is like a blabbermouth who ruins a movie by telling you how it ends. There are some things we don't want to know. Important things.
-- Ned Flanders
Once something's been approved by the government, it's no longer immoral.
-- The Rev Lovejoy
Posts: 7392
Threads: 53
Joined: January 15, 2015
Reputation:
88
RE: The future of AI?
March 28, 2016 at 1:51 am
(March 28, 2016 at 12:57 am)IATIA Wrote: I take it that you are not involved with microprocessors at the machine level.
I take it that you have never tried writing AI before.
(March 28, 2016 at 12:57 am)IATIA Wrote: Microprocessors have an interrupt input that will force it to jump to a subroutine. This subroutine would check for interaction with a human.
How would you write this subroutine that checks with interaction with a human?
Posts: 5356
Threads: 178
Joined: June 28, 2015
Reputation:
35
RE: The future of AI?
March 28, 2016 at 5:12 am
(March 28, 2016 at 1:51 am)Mathilda Wrote: (March 28, 2016 at 12:57 am)IATIA Wrote: I take it that you are not involved with microprocessors at the machine level.
I take it that you have never tried writing AI before.
(March 28, 2016 at 12:57 am)IATIA Wrote: Microprocessors have an interrupt input that will force it to jump to a subroutine. This subroutine would check for interaction with a human.
How would you write this subroutine that checks with interaction with a human?
Visual recognition as in facial recognition and voice recognition seems to be a good start.
Posts: 7392
Threads: 53
Joined: January 15, 2015
Reputation:
88
RE: The future of AI?
March 28, 2016 at 6:57 am
(March 28, 2016 at 5:12 am)pool the great Wrote: (March 28, 2016 at 1:51 am)Mathilda Wrote: How would you write this subroutine that checks with interaction with a human?
Visual recognition as in facial recognition and voice recognition seems to be a good start.
How it would figure out what was on TV and what was real life?
Or a play being acted in a theatre?
Posts: 4196
Threads: 60
Joined: September 8, 2011
Reputation:
30
RE: The future of AI?
March 28, 2016 at 7:04 am
One would not necessarily need anything as complicated as facial recognition. Simple thermal imaging might do.
You make people miserable and there's nothing they can do about it, just like god.
-- Homer Simpson
God has no place within these walls, just as facts have no place within organized religion.
-- Superintendent Chalmers
Science is like a blabbermouth who ruins a movie by telling you how it ends. There are some things we don't want to know. Important things.
-- Ned Flanders
Once something's been approved by the government, it's no longer immoral.
-- The Rev Lovejoy
Posts: 7392
Threads: 53
Joined: January 15, 2015
Reputation:
88
RE: The future of AI?
March 28, 2016 at 7:31 am
(March 28, 2016 at 7:04 am)IATIA Wrote: One would not necessarily need anything as complicated as facial recognition. Simple thermal imaging might do.
Given all it has is a thermal image, how does the AI understand how the human is interacting in the world?
How can it recognise a group of human beings huddled together? Or drowning in a very cold lake? Or has its wrists slit in a very hot bath?
How does a robot recognise when itself is in danger? How can it know the consequences that saving itself will have consequences on other humans?
Posts: 4196
Threads: 60
Joined: September 8, 2011
Reputation:
30
RE: The future of AI?
March 28, 2016 at 12:08 pm
(This post was last modified: March 28, 2016 at 12:08 pm by IATIA.)
(March 28, 2016 at 7:31 am)Mathilda Wrote: Or drowning in a very cold lake? Or has its wrists slit in a very hot bath?
A good programmer can overcome all these difficulties and specifically addressing the above quote, if the robot is doing it's job, the above scenarios should not happen in the first place. And ultimately, in the end, shit just happens.
You make people miserable and there's nothing they can do about it, just like god.
-- Homer Simpson
God has no place within these walls, just as facts have no place within organized religion.
-- Superintendent Chalmers
Science is like a blabbermouth who ruins a movie by telling you how it ends. There are some things we don't want to know. Important things.
-- Ned Flanders
Once something's been approved by the government, it's no longer immoral.
-- The Rev Lovejoy
|
