The “Ghost Internet” Theory

Ben Arthur

Elite Member
Jr. VIP
Joined
Sep 9, 2023
Messages
3,032
Reaction score
1,801
Let me throw a strange idea out there.

What if a large part of the internet you interact with in the next 2–3 years won’t actually be run by humans?

Think about it.

AI can already write posts, answer comments, create images, generate videos, reply to emails, and even hold conversations that feel real. Now combine that with automation.

One person could run hundreds of accounts across forums, social media, communities, and websites.

Accounts posting content.
Accounts replying to discussions.
Accounts sharing links.
Accounts are arguing with each other.

All automated.

Now imagine thousands of people doing this.

At some point, you start to wonder how many discussions online are actually between real humans and how many are AI systems talking to each other.

A Reddit thread.
A Twitter debate.
A product review section.
A forum discussion.

It could easily be 5 humans and 200 AI agents.

The crazy part is that this is not really a future prediction. The technology already exists.

We are entering an era where the internet might slowly become a “ghost network” where a huge percentage of activity is automated.

Which leads to a weird question for anyone doing internet marketing.

If AI is creating the content, posting the discussions, and even driving engagement… what exactly will the human role be?

Curious what others here think.

In 3–5 years, what percentage of the internet do you think will actually be AI-generated?
 
most of the engagement will be people arguing in the comments (as usual)
a few people uploading personal videos with facial recognition proof to validate that its a real video - will get increased traction likely
 
interesting, internet will be full of bot (its already more than 50%)
 
Honestly this is already happening. Most comment sections and forum replies feel noticeably different from 3-4 years ago. The tell is usually perfectly structured opinions with no real personal angle or specifics.
The question is whether it matters. If AI-generated content is genuinely useful and accurate, does it matter who wrote it? Probably not for information. Probably yes for anything requiring trust or real experience.
 
most of the engagement will be people arguing in the comments (as usual)
a few people uploading personal videos with facial recognition proof to validate that its a real video - will get increased traction likely
You’re probably right. Arguments will still drive most of the engagement since people naturally jump into debates.

The verification idea is interesting too. If AI content keeps growing, real human proof like personal videos or live streams, might become a strong trust signal. Authentic content could actually become more valuable.
interesting, internet will be full of bot (its already more than 50%)
Honestly, it already feels that way. A huge portion of traffic, comments, and even social media interactions are likely automated at this point.

If AI tools keep getting easier to use, the percentage will probably keep rising. The real challenge will be figuring out what content and engagement is actually coming from real people.
Honestly this is already happening. Most comment sections and forum replies feel noticeably different from 3-4 years ago. The tell is usually perfectly structured opinions with no real personal angle or specifics.
The question is whether it matters. If AI-generated content is genuinely useful and accurate, does it matter who wrote it? Probably not for information. Probably yes for anything requiring trust or real experience.
Good point. You can often spot it by how polished everything sounds, but there is no real experience behind it. No small details, no personal context.

And you are right, for basic information, it probably does not matter much if AI wrote it. But when it comes to reviews, advice, or anything where trust matters, people will still want real experiences from real users. That is where human content will probably stand out more.
 
You're already right. I've seen entire review sites where it's clearly AI talking to AI in the comments. The tells are getting harder to spot.

The economics break everything though. When you can generate 1000 posts for $5 in API costs, content becomes worthless. Verification becomes everything.

Reddit's already 30-40% bots in some subs. Twitter's worse. In 3-5 years I'd guess 60-70% of public internet content will be AI generated.

Real question is do people care? For pure info probably not. For trust and transactions absolutely yes.

We're gonna see proof of human mechanisms become standard imo
 
You're already right. I've seen entire review sites where it's clearly AI talking to AI in the comments. The tells are getting harder to spot.

The economics break everything though. When you can generate 1000 posts for $5 in API costs, content becomes worthless. Verification becomes everything.

Reddit's already 30-40% bots in some subs. Twitter's worse. In 3-5 years I'd guess 60-70% of public internet content will be AI generated.

Real question is do people care? For pure info probably not. For trust and transactions absolutely yes.

We're gonna see proof of human mechanisms become standard imo
Exactly. Once content can be produced almost for free, quantity will overwhelm quality. That makes verification the new currency online.

I agree. Trust will become the key differentiator. AI can provide information, but for anything involving real decisions or transactions, humans will still matter. Proof of authenticity is probably going to become the standard across platforms.
 
Wouldn’t be shocked if half of it ends up bots talking to bots in a few years.
 
Wouldn’t be shocked if half of it ends up bots talking to bots in a few years.
Honestly, that seems very likely. Once AI can generate content at scale, a lot of threads and discussions could just be bots interacting with each other. The real challenge will be figuring out what comes from humans.
 
Let me throw a strange idea out there.

What if a large part of the internet you interact with in the next 2–3 years won’t actually be run by humans?

Think about it.

AI can already write posts, answer comments, create images, generate videos, reply to emails, and even hold conversations that feel real. Now combine that with automation.

One person could run hundreds of accounts across forums, social media, communities, and websites.

Accounts posting content.
Accounts replying to discussions.
Accounts sharing links.
Accounts are arguing with each other.

All automated.

Now imagine thousands of people doing this.

At some point, you start to wonder how many discussions online are actually between real humans and how many are AI systems talking to each other.

A Reddit thread.
A Twitter debate.
A product review section.
A forum discussion.

It could easily be 5 humans and 200 AI agents.

The crazy part is that this is not really a future prediction. The technology already exists.

We are entering an era where the internet might slowly become a “ghost network” where a huge percentage of activity is automated.

Which leads to a weird question for anyone doing internet marketing.

If AI is creating the content, posting the discussions, and even driving engagement… what exactly will the human role be?

Curious what others here think.

In 3–5 years, what percentage of the internet do you think will actually be AI-generated?
I believe that in 5 years the percentage of the AI generated content and actions will be at 80%.
 
Back
Top