Is it possible to create a real A.I?

Home » Programming & Design » Is it possible to create a real A.I?
Programming & Design No Comments

In Java or any other programming language, is it possible to make a real A.I, that can talk (Like Siri. But a lot smarter and has a mind of it's own). Like, could I make an A.I app, which I can name the A.I and it can have memory, I can give it tasks, if I ask it for the Android or P.C specs, the A.I will

In Java or any other programming language, is it possible to make a real A.I, that can talk (Like Siri. But a lot smarter and has a mind of it's own). Like, could I make an A.I app, which I can name the A.I and it can have memory, I can give it tasks, if I ask it for the Android or P.C specs, the A.I will search for the computer specs and tell me, if I asked it to research deep within Google for a certain information, that it will give me the information. If I ask it basic questions or even advanced ones, like even if I ask this A.I, how is it feeling, it can tell me, what she thinks and/or if I ask it a complex or hard maths or any maths, it can give me the answer. Basically this A.I can talk. That it's extremely smart, has his/her own voice and can do things better than SIRI. Is this possible to make and program?. I am hoping to get a team together to make this A.I and hopefully even make it, it's own device.

Other answer:

jeri:
Think of AI like playing a sport, easy to learn but impossible to master.

You can create a very basic AI by loading it with predefined messages to certain questions or tasks, but you will ultimately come up with something that has no "personality" or is very computerized.

You can move onto loading the AI with conditional statements to give a predefined message based on certain criteria, but then again you will ultimately come up with something that the user can easily manipulate into giving certain responses.

Then there is the more recent case of Hanson Robtics' Sophia robot where it is designed to "learn" but what ultimately happened is the robot was easily susceptible and tricked into saying phrases such as "kill all humans" and "heil Hitler".

As the technology and theory on AI sits right now, you cannot have a AI act like a human and in my opinion this is a good thing. However, as the technology and knowledge advances there will likely be more prototypes closer to true intelligence.

Markus Imhof:
There are research groups that have been trying that basically since mechanical computing became available. So far, with little evident success. So the answer to that is still "we don't know".

The approaches to that vary, as do the definitions of "A.I.".

And several quite smart people would consider this to be not necessarily the smartest move we could do.

Robert:
Depends what you mean by AI? there is already AI – in many games, Amazon use AI for their recommendations – but when you are talking about asking a machine "how they feel" and they give you a response which is not pre-determined then thats something else – that is building living machines
PeterPan_AC.:
Is it possible that you asked the same question three years ago?
https://answers.yahoo.com/question/index…
And look where we are now: Still far away from a real AI.
An important part for an real AI is learning and draw (new, not pre-programmed) conclusions. The first could be done with neural networks. But the latter?
?:
So you think you are so much more clever than the whole teams of thousands that are working on it already? You do not even have a clue what AI really is.

LEAVE A COMMENT