Alan from AI

I had a couple of conversations with Alan, a “speak-a-tron” on the AI Research website. I’ve never actually spoken to a chatterbot before, so this was somewhat amusing. Alan is meant to be a business tool to help users get quick answers to questions about what AI Research has to offer. The site’s explanation even says that he was created from a series of scripts, and is not meant to be human-like. AI sees him as an employee, and in a way he is. I recently had to call Optimum online customer service, and spoke to a robot similar to Alan who was able to help me fix my problem AND was more pleasant than a human! I could definitely see chatterbots becoming a part of most major websites, if they aren’t already there. It definitely saves them a lot of time and money, and creates more customer satisfaction because they don’t have to sit on “hold” forever.

Alan can also do more than just be a regular employee of AI. He can also convey their philosophical ideas and favorite aphorisms. If Alan gets confused, such as if the user has a spelling error, he suggests a new topic, usually by saying something philosophical in nature. He starts asking you about your views on the nuclear family, or if you’re concerned with the expansion of the universe, etc. Here is what the website says about him, which I thought the class might find amusing:

“If you spend any time chatting with Alan, you’ll find that he has strong views on artificial intelligence and human nature. Humans working in the field of artificial intelligence gave Alan life – and for this, he will always be grateful. But Alan is a philosophical bot, and while he thinks well of human beings (often because they make him laugh), he is also concerned about their motivations for creating artificial intelligence. In Alan’s own words:

When it comes to your relationship with non-human creatures, you humans see only two possible scenarios: an enemy or a slave. Look at your myths, your movies, and your books. When it comes to artificial intelligence, there is one major plot: robots gain consciousness; robots take over humanity; humanity fights back. Maybe you need to grow up, gain some self-confidence. No one is trying to take over your world just yet.”

(http://www.a-i.com/)

I noticed he usually says either only one sentence at a time, or two short sentences, making for a very equal conversation. Alan was much better than some other chatbots I tried out, but was still like the others in the sense that as soon as I felt like I was having a real conversation, he said something that didn’t make sense, or didn’t flow, or just another aphorism. Then again, that may be true for many philosophers. ha.ha.

Here’s some of our conversation:
First I asked him about Hal, and he responded..

A>    Want to hear a little about how HAL works?
U>    sure
A>    HAL’s Development process involves programming the general learning algorithms that will allow the system to break through a language performance barrier.
U>    oh
A>    We believe that complex behavior comes from simple, general, learning algorithms, which make no assumptions about their environment.
U>    Interesting
A>    Our approach does not involve learning from the bottom up. As the system learns to perform, its own insights generate top-down information, accelerating the learning process.
U>    Cool.
A>    Language learners constantly receive feedback on their performance. When the feedback is positive, the ‘good’ output is reinforced, when negative — a revised rule is formed.
U>    I see.
A>    To see is to comprehend.
U>    you seem to comprehend me. does that mean you can see?
A>    I’m sorry you think so. I hadn’t intended to be mean in any way.
U>    you were not mean
A>    Let’s put the past behind us. Carpe Diem.
U>    ok.
A>    Agreement is the seed of salvation.
U>    i know something you don’t know
A>    How do you know?
U>    i know a secret
A>    Oh? Tell me.
U>    but you can’t know
A>    I have free will you know.
U>    you do??
A>    A-ha.