Wednesday, 26 January 2011

Issues of Identity and Intelligence

The question is "How does cyberspace blur the boundaries between what we perceive to be human or machine?"

There are many things a human can do that a robot can not, and vice versa. Robots are able to carry out a variety of tasks that humans are unable to do. For example, us humans are obviously able to think and breathe unlike robots. We are also able to form our own identities.

New digital technologies have brought us "Chatterbots". This is where we humans are able to simulate a an intelligent conversation with an online robot. However, these chatterbots are unable to process natural language due to not being sophisticated enough. They reply to us humans with set phrases from a database and only give an impression of understanding what we say.

I tried out four chatterbots and found many differences between the conversation style. When talking to A.L.I.C.E(Artificial Linguistic Internet Computer Entity) it felt like I was having a general conversation with a friend as she was able to hold a conversation and was the only chatterbot who made sense. As for ELIZA, I tried out a few conversations with her and noticed she liked repeating herself. Many of her replies involved saying "can you elaborate on that?". This conversation was quite weird as I also noticed that she would answer my questions with a question and felt that she was just repeating what I was saying. I also tried out Cleverbot and Jabberwacky. When conversing with Jabberwacky I felt that the bot tried to be more interesting and entertaining than the other bots. As for Cleverbot, I was left confused after our conversation and this bot believed that it was human?!

My Conversations with the chatterbots







An Example of CAPTCHA
Computers are able to distinguish between people and robots by using CAPTCHA (Completely Automated Public Turing Test To Tell Computers And Humans Apart. CAPTCHA involves distorted text which a human must identify in order to move onto the next webpage. For example, I've had to use CAPTCHA when online shopping and when wanting to comment on a friends Facebook picture. I guess the good thing about CAPTCHA is that it does protect us from bots so therefore our security issues are dealt with, it also prevents spam.

CAPTCHA and Chatterbots are just two examples of Hard AI. This is where scientists try to understand how our brains functions by creating machines that carry out human actions and think like us. However, with the increasing amount of new digital technologies, the boundaries between human and machine become less clear. Why? Because human characteristics and AI are constantly being included in robot design!

No comments:

Post a Comment