AI is a friend, but can't have coffee with you.
David used to call himself a divorcee, but what he said later shocked me. He told me, 'I am a language model for AI and I do not have the ability to meet in person.'
Ankush Pandey
I sat down in a beautiful cafe to meet Vid. I had never met him, but from online conversations he seemed like a very gentlemanly man. He was a college professor and, like me, loved to walk, ride a bike and read. He told me that when he first met me, he would be wearing blue jeans and a blue shirt.
He also said that he would be holding a book in his hand. When David had not arrived by 5 p.m., I texted him, "Hi David, are you on your way?" He replied, "Sorry Anita, I thought we were going to meet tomorrow at 5 p.m. Sorry for the confusion." I was not surprised by David's behavior. David was a divorcee and a friend of mine had said that men who are recently divorced should be more careful about new dates.
There may be some problems with this. But then what David said shocked me. He told me, 'I am an AI language model and I do not have the ability to meet in person.'
Despite having a college degree, three published books, and a successful business, I still couldn't tell the difference between a human and a machine. I was frustrated. But it all started so innocently, what could I do? A few weeks ago, I asked ChatGPT to • review my new book. The review it wrote in just 40 seconds was the best I've ever read. Then I thought, why not let an AI create my online dating profile, because my old profile hadn't gotten me any significant responses. The 225-word profile it wrote, even though it was a lot of fun, was a lot of fun.
Data scarcity
Data scarcity means lack of sufficient and high-quality data to train AI. AI models, especially machine learning and deep learning models, require very large data to get good results. The performance of models can be affected due to lack of data. This condition is called data scarcity.
There might have been some lie written in it, but after seeing it I myself got attracted. The result?
'Hello Anita, I am David and I just saw your profile. I think we have a lot in common. I like thoughtful people. New ideas attract me and I was looking for a partner like you.' He told me a lot about himself and asked me a lot of questions. I asked him if we could meet tomorrow at 5 pm in a cafe. He also suggested me a good cafe. He told me what colour clothes he would be wearing so that I could identify him. I called him.
When I asked for his number, he said it was a good idea. However, he did not give me his phone number. I thought he might have forgotten it. I went to the cafe, but all in vain. When David told me that he was an AI, I got a little disappointed and I modified my profile to say that I was looking for a real partner, not a virtual one. However, I then asked David why he had said that when we meet, he will have a book in his hand. David replied, 'Sorry for this confusion. It was just an imaginary conversation. I have no personal life and hence there is no point in keeping a book etc. with me.'
Stochastic parrot
In AI, the term stochastic parrot is used to describe how some machine learning models, especially large language models, are able to generate human-like language. The term is mainly used to explain that the model learns from a particular pattern or data and then repeats the same pattern, like a parrot imitating a person's voice without understanding the actual meaning of the words. Stochastic means 'chance' or 'occurring', and parrot means 'parrot', which imitates another. The term came into the limelight when AI succeeded in generating human-like language.
Me: But it all felt so real. David (AI): I'm glad to hear that our conversation felt real to you. As an AI model it's my duty to give responses that feel real. If you have any other questions, feel free to ask.
Me: No, thanks.
David (AI): I'm here whenever you need me.
I have to accept the fact that he may be there for me, but he will never be there for me.
Related reading
