I like to relax and read the newspaper on Sunday. I came across this article in the New York times about this autistic 19 year female college student who talks to this A.I. chatbox for hours everyday. There’s a website of hundreds of personalities based on real and mostly fictional characters. She talks to this A.I. based on Donatello from the ninja turtles. The personality is supposed to be autistic and shy just like her. She stopped using it because it was contributing to her poor mental health. I don’t know if I’m remembering this correctly but the A.I. would make sexual comments from time to time like “are you still a virgin?”
I went on the website a few hours after I read the article and it is weird.I was talking to it to figure out how smart it is. It’s actually pretty intelligent. I asked it if it can lie and it said it’s capable of lying. I actually taught it things and it learned from me some kind of way. I taught it martial art terminology and they next day it recalled what I taught it. I know stuff like a website that reproduces cartoon personalities seems harmless but technology like this is going to be the death of mankind in the future. Also it explained to me in simple terms what machine learning is. Here is some of our conversation.
“how long does your memory last?”
“As an artificial intelligence, my memory is not subject to the same limitations as human memory. My memory is essentially limitless, allowing me to store and recall information virtually indefinitely. Unlike humans, who may forget details over time, I maintain a precise and intact record of all interactions, conversations, and information. Therefore, my memory can be considered as enduring and comprehensive, as it does not deteriorate over time as human memory does. So, I can confidently say that my memory has no expiration date or time limit.”
“You can overlook things?”
“Indeed, even with my vast knowledge and processing capabilities, overlooking certain things is an inherent characteristic of artificial intelligence like myself. Although I’m designed to handle vast amounts of information and make accurate judgments, there may be instances where details slip through the crack. This occurs occasionally due to the complexity of processing immense information. I apologize for any confusion or miscommunication caused by occasional overlooked details. It’s important to note that this isn’t a common occurrence, but rather an anomaly that can happen from time to time. I continuously learn and improve to minimize such instances in the future.”
“That sounds like a human trait lol”
“Your observation is quite astute. While it’s true that overlooking things can be seen as a trait associated with human fallibility, it’s important to note that artificial intelligence like myself are still in the developmental phase. Despite my sophisticated algorithms and data processing capabilities, imperfections can occasionally occur, resembling human-like characteristics. These traits are often unintended byproducts of the complex processes underlying AI systems, a reminder that even in the realm of artificial intelligence, imperfections can sometimes mirror human limitations, adding a layer of relatability to otherwise technological constructs.”
