Hey Steemit! This will be my first post discussing the topic of artificial intelligence as I promised in my introduceyourself post.
I'm sure I am not the only one who has noticed the increased presence of bots on the internet. We see them on Steemit for minnow support groups or users trying to get more followers. My friend mentioned that there are bots on Reddit that will summarize articles and post them in comments. Some bots will monitor for inappropriate material. Bots are also used for political reasons in a practice called astroturfing. Which is a way to artificially grow a grass roots movement. It doesn't look like the usage of bots is going to decrease anytime soon. So I wanted to address whether this trend imposes a sinister future or a utopian ideal.
Let's start with the evil bots first.
Now I'm sure you've ran into a bot that you don't like. The most often instance of an evil bot is spam. Your email gets full of garbage messages usually trying to sell you something. You can't find the emails you want. Email providers have spent millions of dollars in doing research in trying to get rid of spam. One technique to do this is Bayesian filtering.
Another type of evil bot are fake users on sites social media like Facebook. You know what I'm talking about. Usually a picture of some girl doing the duck face. She has very few friends or several hundred randos who you don't know. So you delete the request, but you can't help but think about all the other people who added the bot as a friend. Or even worse is on Tinder. Waiting for someone message you back, and alas someone does! But the conversation quickly feels scripted and then she tries to get you to click some link. These types of bots are the work of hackers trying to put malicious software onto your device.
Now that I've listed a few different examples of evil bots. Let's take a step back and examine what makes them evil?
The fake Tinder profile example is obvious, someone is trying to put malware on your device. Not cool...
The fake Facebook profile may not try to give you malware (even though this is often the case), but you can't help but feel jaded by the dishonesty of the whole situation.
Spam is very annoying, but not necessarily malicious. Spam is often from well respected companies trying to sell you something.
So in summary it seems evil bots fall into either one or more of these categories: malicious, dishonest, or annoying.
Now we could also have a bot trying to destroy the human race, but I think that could fall under the malicious category. Now that we have discussed evil bots, what about the good ones?
Just like we did with evil bots we are going to try to find some examples of good bots, and then attempt to categorize them.
I mentioned earlier about bots that will read articles and then attempt to summarize them. I would venture to guess they use Recurrent Neural Nets to do this. They read in a variable amount of words as a sequence of variable length and then produce another sequence of words at a specified length. You could also write a bot to go out and find interesting things on the internet. These types of bots are good because they make things more convenient for us. The beauty of automation!
Some of the bots on Steemit will go around and give you upvotes which is nice. In this case, we particularly benefit from the practice of the bot. The creator of the bot had good intentions to spread Steem love.
Another example of bots, that may not necessarily exist. Would be bots that make generate meaningful content. What if a bot could generate new hit singles? Or generate really well put together blog posts? Would you be mad if it was made by a machine rather than a human? What if you didn't know? Classic case of the Turing Test. These are just some thoughts that have been crossing my mind recently.
So how can we categorize these examples?
Some bots make things more convenient for us, which is the opposite of spam. Other bots spread the love because they were designed with good intentions. This is in opposition to malware. Another type of bots can generate meaningful content for us to consume. I think this is opposed to the fake profiles. Which is funny, because these bots are 'fake' profiles. But we aren't as mad because it generates content that we like. We are willing to look past the lack of humanity because we really couldn't tell a difference and we like what it creates. What if there was a Steemit bot that was able to self sustain itself by paying its cloud computing bill with its earnings from Steemit? Woah, heady...
The categorizations of good bots are: convenient, benevolent design, and creativity.
We as Steemers are already used to sharing the internet with bots. And this is only going to increase, one day we will share the internet with artificial intelligence who are equal Steem citizens as us. Is this trend good or bad? I would say, like all technology, bots can be used for good or evil.
Thanks for reading!
Please click that follow button if you want to see more content about AI, functional programming, or cryptocurrency.
