Dead Geekhack theory?

[blondie]

If you've not heard of the dead internet theory, it is the notion that most of the communication on the internet is by automated software. I am starting to wonder if this has come to geekhack. Just look at this thread, for example:

https://geekhack.org/index.php?topic=99897.0

I get the feeling that those new accounts, with their vacuous replies that don't even entirely make sense come from some automated software.

[Hugo_mungus]

Wow! This information you’ve uncovered is surprising and revelatory: you’ve shown that robots are everywhere on GeekHack! I’ll use these tips in the future to get a deeper understanding of who- or what- is discussing keyboard goods and services. I only want to talk to humans; no bots , please!

Great work, sleuth! You’ve saved the day.

[Findecanor]

A few of us are trying to report as many of those as we see to the mods, but you can't visit all threads.

[Rob27shred]

While a little hyperbolic there is some truth in Dead Internet Theory. Bots & AI have definitely invaded every corner of the internet & will most definitely be on any forum type site like GH. However the bulk of posts & responses are still from actual humans here. Also thankfully just like you pointed out it is pretty damn obvious which accounts are bots & which are human. While D.I.T. my eventually become a self fulfilling prophecy, especially with AI progressing as fast as it is I don't think we're there quite yet.

[Findecanor]

It hasn't always been obvious. Many posts from SPAM-bots have been a new thread with a repost of an existing OP.
Sometimes they haven't made sense in their context, but when they have they have often got replies from other users.

The biggest red flag is if it has been a new user posting their first post outside the New Members subforum (but sometimes spambots post there too).
Most bot accounts have had very generic usernames in full lowercase (auto-generated from some formula, probably).
The posts have often been harvested from Geekhack or Reddit. Any reference to Reddit is a red flag.

[Piperaceae]

Between the bots, constant 403 errors, and multi-week delays in getting mod approval for GB posts, it's hard not to come away with the impression the site is in free-fall.

[blondie]

I haven't had connectivity issues, but maybe that is region specific. Also, I haven't wanted to post anything that requires a moderator's approval. I think they are looking for new moderators.

Apart from that, I am glad the bots and automation software is still noticeable. I have wondered how we will communicate if it ever gets to a point that we have a difficult time truly knowing what we are talking to. Will we need some form of certification authority which verifies that we are human beings in the real world, and that certificate is attached to all of our posts? But of curse someone could sell (or lose) their cert to someone else who in turn uses it in some software spamming tool.

In other words, at the extreme end with software that is perfectly indistinguishable from humans, we would need a whole infrastructure built around validating actual humans and their messages lest we just chat with bots 'til the end of time.

And that also comes with a ton of issues, including one of privacy and anonymity. How can we continue to communicate anonymously online if we have to have some sort of an equivalent of a notary public giving a stamp of approval on our huamanness?

Anyway, lucky thing we're not there yet.