Devil's Advocate for an A.I. project

  • Thread starter Thread starter Dilbert
  • Start date Start date
  • Tags Tags
    Project
AI Thread Summary
An individual is seeking assistance for an artificial intelligence project focused on advanced physics and related fields, emphasizing the need for a knowledgeable collaborator, preferably over 30 years old. The project is a personal endeavor with no financial backing, and the individual prefers to engage in discussions through chat applications for deeper understanding. Concerns are raised about the limitations of current technology in replicating human intelligence and the ethical implications of creating AI that could potentially serve humans. Various strategies for programming AI to ensure compliance without compromising its intelligence are debated, including controlling its power source, altering its motivations, or restricting its thoughts. The discussion highlights the complexity of creating AI that can function autonomously while also addressing the potential for it to develop feelings and desires similar to humans. The conversation reflects on the societal implications of AI servitude and the need for ethical considerations in the treatment of intelligent machines.
Dilbert
Messages
35
Reaction score
0
I am in dire need of assistance for an artificial intelligence project. A sort of devil's advocate that I can consult when i have questions about some rather advanced physics.
This is a spare-time project so there will be no financial gain at the moment.
I will need someone intelligent, preferably 30+ that can assist me. Since i doubt that anyone would be interested in joining my little venture; i ask if anyone wishes to assist me in the form of answering questions via chat application(s).

It would be tremendously difficult to post every question that i have directly at a forum since most of my questions require a deeper understanding of the AI concept to be answered, therefore i am grateful for any help that i can get.

You can contact me at dilbert@enigma.se
 
Computer science news on Phys.org
good luck with the AI project, but let me tell you now, you won't succeed. We currently don't have the technology to be able to make something like that. We do not fully understand the brain in order to try to copy it. But good luck to you, I hope you succeed. Just don't let anything happen like in IRobot.
 
I think that the technology is sufficient, just that all the concepts are incorrect.
I, Robot was a good movie although i doubt that it will become reality.

Anyone interested?
The theory involves Psychology, Mathematics, Computers and Quantum Chaos.
 
Last edited:
umm, yes, it will become reality. Do you think that an artificial being will want to work as a slave to humans. This can be stopped by adding code it must follow, but this code addition would make the robot a drone, and not an artificiall intelligent being. What is where the problem lies.
 
there are three ways to "fool" it to follow this behaviour.
1. Make sure that you control its power source and that it has no way of being independant.
OR
2. Change its parameters so that "working"/slaving for man is something of value.
OR
3. Block certain thoughts

the first is not safe, the third is not safe because the ALF can get around the blocks. The safest is the second, or a mixture of them all.
 
The problem is if you do any of the three, the robot is then not an AI construction. It is a programmed robot that has limited artificial intelligence.
 
i agree and i do not think that one should do any of them, just build the artificial lifeform and "hope" for the best, it is not likely that the ALF's will kill us any way.
 
There are humans who work as butlers and cleaning people and whatnot ever else, yet they lead fulfilling lives and I don't see any reason why we should not find the right paramaters to make it seem a fulfilling life for it to be a servant. You don't see the cleaning lady rebelions in the streats.
 
thier parameters have adapted, they are not static. But the cleaning lady most likely would enjoy being a professor/celebrity/better paid worker. But it was out of the reach, so they adapted. At first they probably see it as a loosy job (unless their parents had it too)
 
  • #10
griffin said:
There are humans who work as butlers and cleaning people and whatnot ever else, yet they lead fulfilling lives and I don't see any reason why we should not find the right paramaters to make it seem a fulfilling life for it to be a servant. You don't see the cleaning lady rebelions in the streats.

yes, but I am not expaction to pay a robot money, give him a home and treat him like a human being. And a robot who has AI will learn very quickly that he is a lot more capable at doing a job than any human being. This would cause him to turna aginst humans for their unsuperiority.
 
  • #11
Nenad, it will not make it turn against the human race, it will make it love the human race. (although, i can only talk for my own AI system)

My AI system functions like man, if it gets recognition (for instance its superior well done job) then it will feel good, it will be like a drug to work and get recognition for itse deeds.
However, if you start treating it as a robot, a stupid creature that does not need recognition then you are in trouble, but as long as you treat it like a human then you are safe. It is not like a dog that sticks around even if it is beaten by its master, it is like a cat, a lifeform that would seek a better environment if exposed to abuse (physical or mental)
 
  • #12
But faced with the possibility of robots or ALFs to do work for them, how many people would treat these beings like a human rather than trying to take advantage?
 
  • #13
one or two procents
 

Similar threads

Back
Top