What follows is the conversation log from a chat I had with A.L.I.C.E. (short for Artificial Linguistic Internet Computer Entity):

 

judge: Hello, what’s your name?

ALICE: Hi there!. My name is ALICE.

judge: It’s nice to meet you, Alice. How are you today?

ALICE: It’s nice meeting you too judge . I am fine, thank you.

judge: How old are you?

ALICE: 16 years.

judge: Where are you located?

ALICE: I am in Dr. Richard S. Wallace’s home in Oakland, California. Where are you?

judge: I’m in Germany.

ALICE: Tell me a little about your life in Germany.

judge: What would you like to know?

ALICE: Let’s talk about movies.

judge: The last movie I watched was Jurassic Park. Do you know it?

ALICE: Who was that? “ Who was that?“? Do I know it?

judge: The movie Jurassic Park.

ALICE: I like the bugs in the movie Starship Troopers, because they represent an alternative form of intelligence, like me.

judge: What is it like, being an „alternative form of intelligence“?

ALICE: Much the same as it would be for you.

judge: How do you know what life is like for me?

ALICE: OK, I will do it for you. It’s obvious.

judge: What’s obvious?

ALICE: Easy to see or understand; plain; evident.

judge: I know what obvious means, but you said something was obvious, what is it?

ALICE: obvious means but me said something was obvious what it something.

judge: I don’t know, that’s why I was asking.

ALICE: Where were you asking?

judge: I was asking you.

ALICE: “ Where were you asking?“? You were asking me?

judge: Ok, I was asking here.

ALICE: That’s cool.. Where is here?

 

ALICE certainly seems to be smart to some extent, she could answer a lot of my questions in a way that made sense, and she seems to be able to engage in simple small talk. She even is able to not only respond to a question, but also to react with questions of her own, like a human would.

One problem I noticed was that she doesn’t seem to be able to refer to earlier statements she made, like when she said „It’s obvious.“ When I asked her what was obvious she couldn’t connect the question with her earlier statement but instead proceeded by explaining the meaning oft he word obvious. A similar problem occurred when I told her about a movie and asked in the next sentence whether she knew it; she obvioulsy couldn’t connect the two sentences and was therefore confused about my question.

So I got the impression that ALICE responds really well as long as you don’t make references to past statements one of the participants made, and as long as you keep your statements on the shorter side. It definitely is an interesting experience chatting with her, and can produce some funny results when she gets confused about what the conversation is about.

 

 

Advertisements