How to spot a bot

Very recently i had this feeling that a big part of our interaction on the Internet is artificial. In the sense that we are no longer interacting with someone but rather something that is mimicking someone.
Sometimes I feel like even youtube videos are artificially generated in the moment you see them. Or at least that the first couple suggestion given by YT are in some way connected to your subconscious mind, they read it and generate content that messes with you on a subconscious level, in real time.
I can not exactly pinpoint what made me feel that way, but maybe there are some markers that can give away the bots.
The whole topic is over my head and i can't even think of a way to trick a bot.
I also imagine that with the rapid evolution of artificial intelligence it becomes increasingly difficult to distinguish.

My question to the ATS members versed in this area is the following:0ä

Is there a way or some markers to identify if the content is generated by a bot or if the content you consume is actually generated by a human?

Sincerely NC

a reply to: NoConspiracy

Thats a good question. I think on a place like ATS it may be more difficult to discern.

Bots on twitter and instagram can be spotted fairly easy with basic metrics but who knows if that info is accurate.

Its the unfortunate world we live in. Most social media accounts for "famous" folks use bots.

a reply to: NoConspiracy

What if you are a bot and do not know it? What if I am a bot and just want to make you insecure by asking this question?

Would you think my bad grammar is intentional

The thing is, many are so programmed, the term bot... would fit! Is this a test? Did I pass?


Add: Here is something I toyed around a while ago:
www.cleverbot.com...

This thing tries to talk you into believing YOU are the bot. It is quite entertaining.

a reply to: ThatDamnDuckAgain

Exactly the stuff i wonder.
I imagine even bad grammar could be part of a human like bot.

With just enough time and data it could probably imitate almost every individual, that has fed it.

Quite sophisticated this clever bot

a reply to: NoConspiracy

Sometimes I had the feeling this just redirected answers a different human typed.

a reply to: ThatDamnDuckAgain

Yes me too... it also contradict itself...

a reply to: NoConspiracy

I came to that conclusion because it had some really naughty comments.

a reply to: NoConspiracy
Apparently the word poop is one that AI has problems with.

a reply to: Skid Mark

If I understand it correctly it is one that people would use to identify as humans.
Now that the information is out there on the internet this information has probably been assimilated by AI, and it would no longer serve as such a marker...

Sometimes I wonder if the Internet has become one giant touring test

First I had to lookup what you mean with touring test, then I disagreed but now thinking further...

When we test self driving cars on the highways / streets, it is kind of a touring test. And the roads are the wires of the internet, does that make the internet one giant touring test? Kind of I guess, if people are testing these intelligent programs on the internet, like that chat bot.

At least it can be a platform for a touring test so you have a good argument

originally posted by: ThatDamnDuckAgain
First I had to lookup what you mean with touring test, then I disagreed but now thinking further...

When we test self driving cars on the highways / streets, it is kind of a touring test. And the roads are the wires of the internet, does that make the internet one giant touring test? Kind of I guess, if people are testing these intelligent programs on the internet, like that chat bot.

At least it can be a platform for a touring test so you have a good argument

It's actually called the "Turing Test", after pioneer computer scientist Alan Turing. This was supposed to be a means to assessing maturity of artificial intelligence, such that for a human interacting with a computer via a text dialog, could the human ascertain if the dialog was occurring with a real human or a computer? True AI should make that dialog indistinguishable from man and machine.

It *should* be relatively difficult for bots to log into most sites now that use modern techniques like CAPTCHA and tasking the user to perform some basic clicking/question answering to prove they are human. Many sites require such a test for user to submit a thread post, and that usually keeps bots from posting messages.

Usually CAPTCHA test require you to move your mouse in a certain way, or type patterns, or select matching pictures from a set of images.

a reply to: SleeperHasAwakened
Thank you for the explanation, I just typed into my duckduckgo and autocorrect did the rest.

A good example how this can impair our mental abilities, I did not even learn to spell the word.

a reply to: NoConspiracy

I know I should have chosen a better user name.

Perhaps my artwork would help convince some of you...



Nerbot is not a bot, not a bot, not a bot *^µkfhgkdfhbks!!??.

a reply to: ColoradoJens

That's Something of a rabbit hole. What if the bot creator has created metrics that point towards real humans in which case the bot becomes a human in your mind and the human is tagged as a bot, in which case you are now accepting misinformation as the truth. Of course I could be a bot trying to trick you.

a reply to: nerbot
So you say that blinking star and color flashing in the signature isn't you communicating with the AI main computer?

a reply to: NoConspiracy

Bots are still created and run...by humans.

Your point is mute as in almost everything is bot linked and looks human

a reply to: NoConspiracy




- JC

a reply to: NoConspiracy

I guess the use of unneeded repetition, such as using the date despite each post already being dated within ATS itself.

Posting about only one side of a story from readily available sources as if they are not the MSM

Other than that I am not to sure

a reply to: SleeperHasAwakened

It *should* be relatively difficult for bots to log into most sites now that use modern techniques like CAPTCHA and tasking the user to perform some basic clicking/question answering to prove they are human. Many sites require such a test for user to submit a thread post, and that usually keeps bots from posting messages.

With face & voice recognition, OCR and other techniques already perfected, it seems that CAPTCHA could be easily over come by higher level AI. The only thing I can think of is perhaps that level of AI is beyond the budget for what most bots are used for.

a reply to: NoConspiracy
Following that line of logic they know about the OP and know that you're on to them. I hope you're not marked for deletion. Good luck to you.