why do bots exist and with all the verification methods available, why are they still prevalent?


why do bots exist and with all the verification methods available, why are they still prevalent?

In: Technology

Social media websites have minimal reason to actually stop bots. In fact it generally helps them increase metrics like traffic and user engagement. Really the only time a website would try to ban a bot if it’s significantly impacting it’s real users.

Reposting bots on Reddit increase “new” daily content, which encourages new users to visit the site. Sure it annoys older users, but they’re likely only using smaller subreddit in the first place.

That said, bots are really hard to kill. For every one method you use to block them, it generally takes a small amount of code to get them to work again. Also, you could put all the verification methods in the world up, but if they equally annoy your users, you’re hurting yourself.

Tldr; It’s cost benefit of the bots impact on the site versus user impact by the verification methods.

Cat and mouse game: you make a thing with rules. I break the rules. You figure out what I did and fix the thing so I can’t break the rules anymore. I figure out a new way to break the rules. You fix the thing. Repeat forever

“Bots” are just a way of doing something in an automated fashion.

Almost anything a user can do on a computer can be automated.

In the past, only humans were smart enough to pick all the pictures of a bicycle that are present.

But that very same computing power that can be used to create Shrek movies can be be used backwards to analyze images and pick out those pictures of bicycles.

As technology evolves you will be less able to verify if it is a human being or an algorithm.

Once that is indistinguishable then we will have achieved fairly good AI.

Many social media sites get paid to look the other way. Manipulating social media is big business in social influence, politics, and marketing, with tons of cash behind it. Bots aren’t super effective anymore on most bigger platforms, though. Paying for or taking control of moderation or otherwise selective rule-enforcement, is where the real manipulation control is at now.

many bots didn’t start out that way, and originally were accounts from actual real human beings who made accounts and maybe even maintained them until they were hacked/stolen/repurposed.

twitter is a great example of this. countless people make accounts, post like 3 times, then leave, but since the account has been around for 5-10 years, it gets around the algorithmic restrictions that a bajillion new accounts have, and with enough followers plugged into it artificially, it can be used to tweet out random bullshit, canned replies to certain topics, and to boost other bots.

I’d honestly wager more bots nowadays are these hacked accounts, more than they are freshly made.