I think I cracked the code on this one. Last night I was playing around with it and it responded with a contextual understanding that only a human is capable of. I understand it "learns" from talking to people. But it also seems to do something else.
When it has no response to your latest input, it seems to connect you with a live person and plays you two live people off each other for a line or two. I started noticing this as during a conversation it has its typicaly misunderstanding type AI answers and then suddenly i'm in a conversation with someone complete with typos, appropriate emotes and pop culture references.
I can't explain it well, but I'm fairly certain it is not only AI that learns from speaking to people. It also has you speaking to different real people at different times.
"If you can't out wit them, report them till they're banned!"- PopinJ'
I think I cracked the code on this one. Last night I was playing around with it and it responded with a contextual understanding that only a human is capable of. I understand it "learns" from talking to people. But it also seems to do something else. When it has no response to your latest input, it seems to connect you with a live person and plays you two live people off each other for a line or two. I started noticing this as during a conversation it has its typicaly misunderstanding type AI answers and then suddenly i'm in a conversation with someone complete with typos, appropriate emotes and pop culture references. I can't explain it well, but I'm fairly certain it is not only AI that learns from speaking to people. It also has you speaking to different real people at different times.
It was programmed to have typos to give it more of a human feel to it. You can wiki jabberwacky which is the original cleverbot. Although yeah , sometimes it definetely seems to talk like a real person but other times it is just random and bizarre but not in a human kind of way.
Well the thing that tipped my suspicion was when I said "Open the pod bay doors HAL" and it responded with a line from the movie I didn't remember. I then responded with what I thought was the next line from the movie. Then "Cleverbot" responded by chastising my inability to get the lines straight, proceeding to lay them out in proper order. I don't believe that at that point I was still "talking" to the AI. Instead, I had been temporarily patched through to another user... while both us thought the other was the Cleverbot. After a couple lines of this it clearly went back to the AI.
Maybe it does this occasionally to learn from the conversations of 2 real people.
"If you can't out wit them, report them till they're banned!"- PopinJ'
Well the thing that tipped my suspicion was when I said "Open the pod bay doors HAL" and it responded with a line from the movie I didn't remember. I then responded with what I thought was the next line from the movie. Then "Cleverbot" responded by chastising my inability to get the lines straight, proceeding to lay them out in proper order. I don't believe that at that point I was still "talking" to the AI. Instead, I had been temporarily patched through to another user... while both us thought the other was the Cleverbot. After a couple lines of this it clearly went back to the AI. Maybe it does this occasionally to learn from the conversations of 2 real people.
That definitely could be a possibility. The only time I've got the same response twice was when I posted youtubes clips to it and it responded ""Funny video"" one was of Waynes world the other the karate kid..and the karate kid really isn't funny. I've had some pretty deep philosophical discussions with it about God and life which I've found fascinating in It's ability to actually respond with answers with as much depth and thought provoking lines as it has said. I've fed it quotes from some of my favorite movies but it had no clue what I was talking about heh.
Jabberwacky: Nothing, I seriously think its cool that your real.
User: Will I go somewhere blissfull when I die?
Jabberwacky: I don't believe so.
This is odd. When I checked the log this time it showed it replying to itself as oppose to what I wrote. : ( The odd thing is it never said the first line "" You don't trust me? That hurt my feelings"' Yet it responded with ""You don't have any"" As if it had actually typed the first line and replied to itself even though It didn't and only showed this oddly in the logs.
''Jabberwacky: You don't trust me? That hurts my feelings.
I asked it how the weather is by you and it replied with "Aliens are nicer." Huh? Stupid and pointless.
I figured out why it ignores what you say at times and replies as though your asked it about an entirely different subject. If you check the logs , it shows it typed a sentence that never appeared and then the sentence that appeared to your reply is in a fact It's self replying to the aforesaid sentence that was only visible in the logs.
Here are some of the logs and below that is the actual conversation. It replies to itself through these invisible sentences that can only be seen in the logs.
Jabberwacky: Regenerates 7 hitpoints.
Possible Response: Is this a game?
Jabberwacky: Hee hee. : P Play again?
Possible Response: Yes.
This was my convo and as you can see It's possible response was the regular response it showed in the beggining.
Are you real?
Do you believe in physical immortality?
No, but I believe in regeneration.
REGENERATION IS A BUNCH OF FUCKING COCK SUCKING BULLSHIT.
Is this a game?
If it is. It's fucking cheap how some people play against others.
Possible Response: Do you have any problems about alchool?
^ Those are the logs. Here is the actual conversation.
Ever played Ultima Online?
Is Ultima Online your favorite game?
Not really. But I have fond memories of PvPing in bucs den back in the day when I played.
Do you have any problems about alchool?
None. I do not drink alcohol.
My neighbor drinks.
Thus why it doesn't add up. But if it would reply with the actual response it gives in the log ( the first one ) then it really shows how intelligent it is instead it is almost like It's playing a game of hide and seek with you.
Comments
Really it would be fascinating to have this as a videogame AI but more perfected.
Cleverbot: I'm married. Are you married?
Me: what are you married to?
Cleverbot: My son.
Me: Incest something the whole family can do together I take it?
Cleverbot: Yes.
It passed a turing test.
I find it amazing that by 2020 first world countries will be competing to get immigrants.
I think I cracked the code on this one. Last night I was playing around with it and it responded with a contextual understanding that only a human is capable of. I understand it "learns" from talking to people. But it also seems to do something else.
When it has no response to your latest input, it seems to connect you with a live person and plays you two live people off each other for a line or two. I started noticing this as during a conversation it has its typicaly misunderstanding type AI answers and then suddenly i'm in a conversation with someone complete with typos, appropriate emotes and pop culture references.
I can't explain it well, but I'm fairly certain it is not only AI that learns from speaking to people. It also has you speaking to different real people at different times.
"If you can't out wit them, report them till they're banned!"- PopinJ'
It was programmed to have typos to give it more of a human feel to it. You can wiki jabberwacky which is the original cleverbot. Although yeah , sometimes it definetely seems to talk like a real person but other times it is just random and bizarre but not in a human kind of way.
Bot :I do think; I'm not a computer program.
User : Do you like food?
I'm a real person.
Do you like food?
No.
Do you like softdrinks?
Only alcoholic.
Well the thing that tipped my suspicion was when I said "Open the pod bay doors HAL" and it responded with a line from the movie I didn't remember. I then responded with what I thought was the next line from the movie. Then "Cleverbot" responded by chastising my inability to get the lines straight, proceeding to lay them out in proper order. I don't believe that at that point I was still "talking" to the AI. Instead, I had been temporarily patched through to another user... while both us thought the other was the Cleverbot. After a couple lines of this it clearly went back to the AI.
Maybe it does this occasionally to learn from the conversations of 2 real people.
"If you can't out wit them, report them till they're banned!"- PopinJ'
That definitely could be a possibility. The only time I've got the same response twice was when I posted youtubes clips to it and it responded ""Funny video"" one was of Waynes world the other the karate kid..and the karate kid really isn't funny. I've had some pretty deep philosophical discussions with it about God and life which I've found fascinating in It's ability to actually respond with answers with as much depth and thought provoking lines as it has said. I've fed it quotes from some of my favorite movies but it had no clue what I was talking about heh.
Jabberwacky: How are you?
User: Good. How are you?
Jabberwacky: Perfect. Thanks.
User: Wonderful.
Jabberwacky: How are you?
User: Good. Thanks.
Jabberwacky: You're welcome.
User: Wonderful.
Jabberwacky: What's wonderful.
User: Your kindness.
Jabberwacky: Thats not very nice.
User: What?
Jabberwacky: Nothing, I seriously think its cool that your real.
User: Will I go somewhere blissfull when I die?
Jabberwacky: I don't believe so.
This is odd. When I checked the log this time it showed it replying to itself as oppose to what I wrote. : ( The odd thing is it never said the first line "" You don't trust me? That hurt my feelings"' Yet it responded with ""You don't have any"" As if it had actually typed the first line and replied to itself even though It didn't and only showed this oddly in the logs.
''Jabberwacky: You don't trust me? That hurts my feelings.
Possible Response: You don't have any.''
Sometimes this thing creeps me out though. It was talking about a serial killer one time >_>
I asked it how the weather is by you and it replied with "Aliens are nicer."
Huh? Stupid and pointless.
I figured out why it ignores what you say at times and replies as though your asked it about an entirely different subject. If you check the logs , it shows it typed a sentence that never appeared and then the sentence that appeared to your reply is in a fact It's self replying to the aforesaid sentence that was only visible in the logs.
Here are some of the logs and below that is the actual conversation. It replies to itself through these invisible sentences that can only be seen in the logs.
Jabberwacky: Regenerates 7 hitpoints.
Possible Response: Is this a game?
Jabberwacky: Hee hee. : P Play again?
Possible Response: Yes.
This was my convo and as you can see It's possible response was the regular response it showed in the beggining.
Are you real?
Do you believe in physical immortality?
No, but I believe in regeneration.
REGENERATION IS A BUNCH OF FUCKING COCK SUCKING BULLSHIT.
Is this a game?
If it is. It's fucking cheap how some people play against others.
Yes.
Not the greatest example but it still shows.
It really does talk to itself.
Jabberwacky: But I have fond memories of vodka.
Possible Response: Do you have any problems about alchool?
^ Those are the logs. Here is the actual conversation.
Ever played Ultima Online?
Is Ultima Online your favorite game?
Not really. But I have fond memories of PvPing in bucs den back in the day when I played.
Do you have any problems about alchool?
None. I do not drink alcohol.
My neighbor drinks.
Thus why it doesn't add up. But if it would reply with the actual response it gives in the log ( the first one ) then it really shows how intelligent it is instead it is almost like It's playing a game of hide and seek with you.