A robot just passed the self-awareness test

[thatguyjerry]

Isn't concluding that one is capable of speech different from concluding that one is one? If I lost both my arms and someone asked me to throw a baseball I could conclude that I cannot, but not necessarily what or who I am. This seems to be more in line with a robot recognizing its own capabilities, which I believe we have already achieved have we not? I think I've seen things in the past about robots being able to adapt to losing one of four walking legs for instance, so how is that much different than the adaptation exhibited as such? The term self-aware seems to lie solely in the construction of this experiment and somewhat misleading.

[Murdoch]

It's not just about looking like a human. Artificial intelligence these days usually involves machine learning, a dynamic development of intelligence.

 

Aye, my post was related to the fact the Turing Test doesn't account for anything like that it only accounts for imitating human like behaviour. Which is not to diminish how it, and his paper drove things forward. It's just not that relevant in terms of judging artificial intelligence. I was agreeing with you, it's not a great test, but it's significance is not how it holds up as a test.  

 

The second half of his paper, meanders around the potential of "learning machines". Which considering was written in 1950, was seriously ahead of it's time.

[Xorbot]

SOON...

 

 

That's awesome but scary at the same time, I hope this don't leads us to something bad :>

 

[jos]

Isn't concluding that one is capable of speech different from concluding that one is one? If I lost both my arms and someone asked me to throw a baseball I could conclude that I cannot, but not necessarily what or who I am. This seems to be more in line with a robot recognizing its own capabilities, which I believe we have already achieved have we not? I think I've seen things in the past about robots being able to adapt to losing one of four walking legs for instance, so how is that much different than the adaptation exhibited as such? The term self-aware seems to lie solely in the construction of this experiment and somewhat misleading.

To begin with that robot could walk with 3 legs.. 1 will be a redudant one. here they have used deep neural network to achieve this 

[Kamina]

 

 

 

It begins. 

[han_han08]

and i just watched ex machina

[Techhog]

[Delusional Phil]

Can you kill it with fire????

Pretty sure you can't ... An EMP would though ... 

 

This doesn't make it self ware. I think therefore i am. I want to see it make a wrong choice for the greater good, i want to see it explain what is a number, i just want to see if it understands.

Hmm ... Now I feel stupid for not understanding the concept of numbers ... What is a number then a part from a solution to know how many objects there are ... What's the essence of numbers ? ( I'll probably go on Numberphile to see ... ) 

[Dabombinable]

Designing a robot to pass a test of self awareness doesn't make the robot self aware. Test for self awareness are straight up pointless.

[ADM-Ntek]

damn someone has to fight all this negativity so her we have data cuddling with a cat. 

 

[DeScruff]

What about the other robots? Is it possible they also said "Sorry, I know now!" but because they are unable to talk they couldn't?
It seems to me like its just able to detect when new information is added - But does it care where the information came from. Would it care if it's self made noise, or if a programer told it that it could make sound. Is my printer self aware since it can detect when I added more paper/more ink to it? - If the answer to that is No because I added paper, would it become self aware if it had a mechanism that filled it's main feed with more paper on "No paper" warnings?

I guess I'm just saying its easy enough to build a machine that is tailored to "pass" a simple test.
Its advanced in the sense of understanding the question, and able to remember the question asked after gaining an answer.

[Techhog]

This doesn't make it self ware. I think therefore i am. I want to see it make a wrong choice for the greater good, i want to see it explain what is a number, i just want to see if it understands.

You're confusing self-awareness with self-consciousness.

[Dabombinable]

You're confusing self-awareness with self-consciousness.

Self-consciousness = sentient=able to think for itself.

[Techhog]

Self-consciousness = sentient=able to think for itself.

No, you just got three different things confused. Thinking for itself would be reason, sentience is the ability to feel, and self-consciousness is an advanced sense of self-awareness.

[Django]

What about the other robots? Is it possible they also said "Sorry, I know now!" but because they are unable to talk they couldn't?

 

Lol, good answer but still none of the Robots seem to have answered the question.

 

 

Selmer Bringsjord set up a similar situation for the three robots - two were prevented from talking, then all three were asked which one was still able to speak. All attempt to say "I don't know", but only one succeeds - and when it hears its own voice, it understands that it was not silenced, saying "Sorry, I know now!"

 

The answer should have identified the robot and not have been "Sorry, I know now!". Also the robots should have done some diagnostics before answering the question like trying to make a small sound rather than attempting to say "I don't know". On the one robot answering the question the other two could have implied that robot without speaking, for instance a human might point a finger.

 

Not impressive IMHO.

[Murdoch]

The answer should have identified the robot and not have been "Sorry, I know now!". Also the robots should have done some diagnostics before answering the question like trying to make a small sound rather than attempting to say "I don't know". On the one robot answering the question the other two could have implied that robot without speaking, for instance a human might point a finger.

 

Not impressive IMHO.

 

In fairness, I would have been impressed if it said "my hat is white".

[WolfDeville]

To hell with you pussies and your skynet fears. Bicentennial Man here I come!

 

I strongly suggest you folks watch that movie if you haven't already

[Django]

In fairness, I would have been impressed if it said "my hat is white".

In fairness it was a "dumb" test. :lol:

[Ertman]

It sounds more like a fail to me.

Do we actually know if it recognized itself as the one speaking? Now I know means nothing.

Also the robot recognizing that it was itself speaking that is not self awareness. More likely the program was designed to capture the it's response as feedback for analysis, programming in a response to analyzing it's own speech isn't doesn't seem like self awareness to me.

[Guest Strangerbob]

this is bad news

 

Why? cuz humans are so fucking amazing? Apart from like 0.01% of humans, the rest are a bunch of autistic animals, only concerned for their personal material well being.

 

if this development increases the chance of self-awareness surviving in the universe, even if we die - we have done good.

 

The top priority is to kindle and spread the fragile light of consciousness within the universe. No matter the cost.

[lightningterror]

this is bad news

Indeed.

[Guest Strangerbob]

if something is self aware it's only one step away from skynet

 

[ElHomer]

No way that is possible at this moment.

[Doobeedoo]

It's interesting but far from anything that can be called a smart AI cause there are many more tests it needs to be done and passed to actually be considered smart independent terminator.

[Guest]

Don't worry 

 

I'll be back