SilverG33k

A very short conversation between me and a bot pretending to be human, it was ac …

Mon 31 Dec, 2018 12:52 pm

A very short conversation between me and a bot pretending to be human, it was actually a pop up dialog (kinda like a chat with online staff) on a website


Anna: I'm online, my name is Anna,  how are you


Me: *thinking its a human* I'm fine


Anna: Good,  are you a day trader?


Me: nah, are you?   Lol


Anna: Before we go ahead, can I get your mail address, incase we get lost, I can always get back to you


Me: *suspecting Anna is a bot* No way I'm dropping my mail


Anna: Oops.. Can you enter a valid mail address


Me: hahahahahaha forgerrit


Initially I thought Anna was human, until she couldn't figure out if I'm   willing to drop my mail or not


You see, AI can't reason on their own, as a matter of fact, they can't think outside the context of their codes

Comments

dhtml

Mon 31 Dec, 2018 04:01 pm

This is interesting, writing bots is what i am now learning to do

omnipotens

Mon 31 Dec, 2018 04:52 pm

@SIlverG33k You'll be amazed when you see better bots that interact better. They will convince you when you're not dropping the email when they require it. They have been trained to understand your many responses until you finally drop your email address. It all depends on the training the bot has gotten. 

limistah

Mon 31 Dec, 2018 07:34 pm
The best at the moment as I have experienced is computer-human solution. It takes a super computer to identify a cat while a 3-4 year old can solve flawlessly. I am no way disputing against AI, we have Siri etc. But to me, having computer do tasks on their own is a thing for the next century habitants to figure out. Research can be on going to keep the improving, but waiting to have a complete and better solution as we see in SciFi movies, I believe we are not there yet.