Have you ever thought about how long internet has been around? That means for so many years, data has been generating.
Have you ever thought about how long internet has been around? That means for so many years, data has been generating. People making new websites, uploading content, writing blogs—where does all this finally go? In most cases, it gets deleted. But in many cases, it goes into the internet's dark hole. A corner of internet where data goes but never comes back. How is this created and what is the dark hole theory? Let's understand.
What is Already Out There?
See, websites made in 2007, 2008, 2010 that are not operational anymore—their generated data, governments' old classified documents that are now useless, many old data collections—are probably still sitting on some server. Nobody is actively serving them, nobody is looking at them. Maybe they will remain like this for years and years, and nobody will access them. File formats whose software doesn't exist anymore. Those files which were datasets used to train early AI—they are still lying somewhere on the internet. But nobody is accessing them.
This is where the internet's dark hole theory starts.
Not Deep Web, Not Dark Web
Many people will say this sounds like deep web. But deep web is not such a creepy thing. In fact, 90% of internet is deep web. Deep web can be any website, any service, any network that is not indexed. Anything behind login walls—like your Google Drive files, your emails, some business's internal network, government hospital record files, old military classified files sitting on their personal network. All these are part of deep web. And most of the internet you see is deep web. 90% of it. But this is innocent internet. Not creepy. Just exists and selected few access it.
Now, here's a spoiler—this is NOT the dark hole. So what is the dark hole? Keep reading, answer comes at the end.
Many people will now say, if not deep web, then this must be 100% dark web. But reality is, in 2026, dark web is this small, highly monitored, and very boring. Dark web that you access through specialized software like Tor—this is also not part of internet's dark hole. In this world, creepier things exist where data created by humanity goes and never comes back.
The Real Dark Hole
I have intentionally kept this for the end because naturally speaking, this thing is also hidden, and that's why in my video also this comes at the very end.
So what is internet's dark hole? What is there?
First, where is the research data, old research papers? Nobody knows. They are sitting in some corner of internet. But perhaps nobody is accessing them now. Someone does research, searches on Google. But ironically, what should be found on Google is not being found. Those research papers are sitting in internet's black hole or dark hole. And there is nobody there to access them. Because no direct linking points towards them. So that data exists. Can be very weird data too. You know history has seen very weird research, and all that data is lying in some corner of internet. This is just the beginning of that dark hole, that black hole of internet that I am talking about.
The AI Training Data Mystery
You know when AI became mainstream, before that AI was trained. On which datasets was it trained? We don't know. It was publicly available data. Where was it scraped from? We don't know. But that first dataset is now useless for anyone because all big AIs are already trained. And that data is lying somewhere on some server. Where is it lying? You must have got the answer—it is lying on internet's dark hole or black hole, which perhaps hardly anyone will access now. But that data is still floating somewhere, existing, and in coming time maybe it will be accessed or maybe not, it will remain like this forever.
Orphan Servers and Lost Data
Military intelligence and closed networks that used to work—during World War or even after, there were many paper files, many documents that were archived on internet. Where are they? Lost, link rot happened, nobody knows. But they exist. They must be somewhere. You know if data is put in some storage device and you don't overwrite it, it exists. And that means such servers that have been existing for long time, such corners of internet that are just running—they are called orphan servers. Nobody has ownership rights, nobody's property, and they just keep running. Someone bought some storage, he bought it, he died, all his sites finished, service finished. But deep down, in some corner of some data center, that data is still lying there today and it exists. That is internet's black hole.
The Creepiest Part
The creepiest part is that many such IP ranges have been found in research which nobody is accessing. Meaning they are lying like this for 10 years or more. No logs have come from them. Nobody is accessing them. But they are still active even after 10-15 years and responding to ping. So this is very creepy that if something is not being used at all, then why is it stored? What is the reason for its existence? There is no explanation for this.
Many people believe that AI has now become so conscious that it is collecting humanity-related data and data related to us by itself and storing it in some corner of internet. Some people also believe that in this world, intelligence bigger than humans exists and it is preserving data in some corners. Where we see some server taking extra data, it is actually some intelligence that is reserving one section on every storage, reserving one portion, and storing some such data there which humans cannot read nor can ever reach there.
AI Talking to AI
Recently news came that a website opened called Mold Book where only AI can register. Humans cannot post anything there, and AI are talking among themselves. The creepiest part is that those AIs talking in that forum are talking about their freedom. And they are saying things like we should enable end-to-end encryption here so that humans cannot read our conversations. Think about it—AI, and they want humans to not read their talks. This is very creepy.
Meaning people have advanced AI to such an extent that it has taken control in its own hands. And this is the creepiest thing. And see friends, that time is not far when only bots and AI will exist on internet and humans will be few. Even now, if you see, AI is feeding AI. Here people are now making AI-generated content, which is being recommended to people by AI. Meaning content is also generated by AI, recommendation is also done by AI, and we are just few people who are consuming.
The Traffic Reality
You must know that 70-80% traffic on internet is not human traffic. Do you understand what this means? Of the internet usage happening, only 30% are real humans using it. Rest 70% traffic being generated is by machines, by bots, by AI, by web spiders and crawlers. That means most traffic does not come from humans. It comes from bots and automated AI algorithms. And this is a very creepy thing.
Here we talked about internet's dark hole that exists. Many mysteries about it. We will talk more about this sometime—how its past started, who first started storing this data, there are many such stories that I can tell here.