The Next Death of the Internet
When the Internet started, it was simply a way to share text. Like a telephone but with writing instead of speaking. Not that big of a deal.
In theory, anyone could join the internet if they had the means but at this point only top scientists had the means. Computers were still new and expensive and the internet was designed as a tool for scientists to efficiently share studies and reports with other scientists. This made the internet a serious, information-rich space where smart people discussed important subjects in a rigorous way.
But then the internet died.
The Graphical User Interface (GUI) introduced windows, icons, menus, and pointers to make internet navigation more intuitive and flexible. Instead of memorising and typing in clunky text commands for every interaction, you could move shapes and symbols around the screen.
This also allowed for images, and eventually audio and video to become part of the internet's media landscape. Instead of dry text and unwieldy bulletin boards, the internet could now become a fun and interactive playground of shapes and colours.
This, along with improvements in the speed and cost of computing made the internet-enabled home computer a commercially viable technology. Subsequently, the serious scientific discourse that had defined the internet before was drowned out by sports news, chain emails, online shopping, and pornography. The internet became something unrecognisable to the scientists that invented it.
This was the first death of the internet.
I’m Ben Fleming, and if you’re a founder or an entrepreneur or a scientist or a forward thinker of any kind, I want to partner with you and use my expertise in health and fitness to help you become the person you want to be so you can build the future you want to see. You and I can build a better world for everyone forever together. Let me know what you’re working on!
With so many people gaining access to the internet, so much content was being added that it became impossible for the average user to keep track of where everything was. We needed search engines to sift through all the stuff and find something useful. To begin with there were lots of ways to search online but one big player emerged:
Google.com
Google had a simple design that was unusual for the time. Most sites were filled with lots of colourful text and gifs displaying news and features. Instead, Google offered an almost completely blank page with a single text box. Just type in whatever you were actually looking for and a page of results appeared, helpfully ranked in order of relevance.
This made Google an easy first point of contact for the internet so it became the default homepage for many. With this tool, the internet was “the world at your fingertips” and anyone could find anything they wanted at the click of a button.
Then, in the year 2000, Google launched AdWords. This put advertising links at the top of the search results page based on what the user had typed into that simple text box. If the user clicked the link, the advertiser would pay a small fee. This model was effective because it both targeted specific keywords and directly tied the cost of advertising to the engagement of the user.
To begin with, this made for a better internet experience but quickly things got much worse. Unlike a library of books or a collection of artefacts, the internet is not stable. New things are being created and updated every single day. This means internet creators were able to respond and evolve their content to match the way it was being searched for.
Clickbait became the norm. Higher quality content that was less clickable got buried under a wave of shocking headlines and sensationalist claims that attracted clicks. The promise of a democratic, rich, information superhighway was lost amidst low-quality, disappointing clickbait media. The internet was dead once again.
Around the same time that AdWords launched, another new form of internet engagement was burgeoning. Like the network of scientists who shared their work online, people were using platforms to share their personal lives with friends, family, and acquaintances. This became known as “social media”. Friendster launched in 2002, MySpace and LinkedIn launched in 2003, but the biggest, most profitable social media company launched in 2004:
Facebook.com
There are people everywhere with interesting and intelligent things to share and sites like Facebook made it easy for anyone to publish their work. However, not everyone is a top scientist working on world-changing research and technological advancements.
When you make it easy to post anything, you’re less likely to get high effort creations. Just because it’s possible to share your most important work doesn’t mean you’re not going to instead share complaints about your neighbours, baseless conspiracy theories, or selfies that make you look like you’re having a better time of life than you really are.
Rather than being important and thought provoking, designed to serve the viewer, the average social media post is trivial, mundane, and self-serving. Instead of helping people to collaborate and interact as intended, social media sites like Facebook created clear paths to vanity, envy, superficiality, and conformity, amplifying the worst aspects of life in a society.
Social media was meant to make us connect more deeply with the people around us. Instead it made us feel alone in an endless shallow sea.
And it killed the internet again.
While alienation arises in part from the emptiness and repetitiveness of the content we see on the internet, another big part of it comes from the powerlessness we feel as we are unavoidably bombarded with so much redundant stuff. The shear volume of it means we struggle to discern between our own thoughts and the hypnotic echos of yet another captivating meme.
Social media platforms tried to improve the quality of the endless posts and adverts by introducing algorithms; use maths and code to show more of the things you like and less of things you don’t by measuring your interactions with the media.
But it’s not easy to measure what you actually like…
Something you can’t stop watching might still be something you hate.
Something you interacted with might be something you never want to see again.
Something you like in isolation might not mean you want to see more of the same. In fact you may explicitly want to see things that are totally different to what you have liked and interacted with before.
But algorithms can only use measurable data. So they can only feed back more of what we show measurable interest in.
Like, Subscribe, Share, Bookmark, Comment, Follow, and watch all the way to the end!
Mass adoption of personal computing, targeted advertising, algorithmic feeds, excessive user-generated content, and social degradation collectively inform Dead Internet theory.
This is the idea that the majority of the content we see on the internet is not created by living people we can talk to in the real world but fake software accounts with automated scripts used to deliberately manipulate the population with misdirection, distraction, and propaganda.
People have been talking bout Dead Internet Theory for years as a paranoid fringe conspiracy. But there are reasons for us to start taking it more seriously.
After taking over Internet search with various advertising models, Google used a portion of its huge revenue to fund further research and development in the field of computer science. In 2017, a researcher named Ashish Vaswani, working in a team called Google Brain, published a paper titled “Attention is all you need”.
In the abstract, he writes:
“We propose a new simple network architecture, the Transformer, based solely on attention mechanisms …Experiments on two machine translation tasks show these models to be superior in quality while being more parallelizable and requiring significantly less time to train… [at] a small fraction of the training costs of the best models from the literature. We show that the Transformer generalizes well to other tasks by applying it successfully to English constituency parsing both with large and limited training data.”
This new software architecture, originally designed to translate text from one language to another, was not only cheaper and faster to run than other models, but also showed a more general ability to interpret the order and meaning of words in a piece of written text.
In 2018, the Generative Pre-trained Transformer was created by OpenAi, based on the transformer model proposed by Vaswani. In 2019 this model was further improved upon and could generate many paragraphs of readable text based on a short prompt. This improved model was called GPT2.
The GPT model was then made commercially available after being improved again, allowing anyone to quickly and cheaply generate many pages of text. And now in 2024, OpenAi has made their latest generative transformer model (GPT4o) available for free to anyone who wants to use it at their website:
ChatGPT.com
This is the beginning of the next death of the internet. Models like ChatGPT take just seconds to create pages and pages of coherent text, usable code, or even a convincing image. The output is often indistinguishable from the kind of empty, trivial, misleading, and alienating content that has characterised the internet since it became available to the public. If we were already somewhat suspicious of the motives behind the internet media we consume, strong skepticism must now become the default, if not total mistrust.
The internet of the future is often depicted as a virtual cyberspace of simulated environments and interactive experiences. This might seem like the next logical step, but the internet of the future won't just be a more mature version of the internet of today. It will be an entirely new creation. Because old internet is about to die. The models will only get better and the problem will only get worse. People will soon learn to simply stop trusting anything they see online, even if it came from a previously trusted source. This will force us to retreat from the digital world to find more reliable and tangible forms of interaction.
When you can’t confirm the validity of a single thing that comes to you via a device or a screen, your concerns must become more local and more physical.
Collaborative communities, private networks, and real life interactions start to look much more desirable and rewarding than life in a deceptive computer-generated world, prioritising face-to-face meetings, physical contact, local issues, and personal introductions over global digital connectivity.
Along with the majority of the high-tech gadgets we use actively today, the internet will become invisible. An infrastructure layer only. No more checking email, scrolling social media, taking video calls, and looking up information online. You won’t be able to trust it. Like the sewers underground, the internet will still serve an important function but it won't be a place you want to visit.
What is death?
The internet has persisted for years with the same name, but very few of the characteristics that defined it at the beginning. It has transformed, not through gradual and stable growth, but through major shifts that create upheaval and difficulty, leaving it unrecognisable to the people that once knew it.
This is the only way transformation and progress truly occurs in the real world.
Think back over your own life. You bear the same name as your younger self but you’re hardly the same. At each stage you have been working with the resources you had available to solve the problems you were facing. This created new problems for you to solve and new emotions to experience. But it also made new resources available. Your body is different, your personality is different, and - as a result - your identity is different.
A key distinction between the internet and you is the internet has no attachment to any of its previous identities. It can adapt to meet the present moment at every stage with innovation and energy, experiencing no remorse for what has gone and no desire to do things the way they have been done before.
A scientist might miss the early high-effort internet. An entrepreneur might wish they had got in when the home computer was new. But a smart, hardworking person who is properly forward-thinking must treat the past with indifference. Any identity that dwells on previous approaches, past failings, and missed opportunities is an enemy to be avoided.
The only things that truly die are those that cannot find ways to let go, transform, and live on in new and unrecognisable ways.
We look to the future …
By anticipating the next death of the internet we can seize the opportunities it presents. Human beings operate at their best in immediate, trusted communities with enough continuity to grow and transform over time, collaborating and building with like-minded friends.
If we build in the real world, physically and locally, with the resources available to us now we can create new information-rich spaces where smart people can gather and work, discussing important subjects in a rigorous way and building a better world together.
Do you think the internet is about to die?
What sounds more desirable to you: real-life interaction or cyberspace?
Comment below or send me a message.
My handle is @GymnasiOnUK on Youtube, Twitter, Substack, and Instagram.
Like, Subscribe, Share, Bookmark, so you don’t miss the next post and take a look at Gymnasion.net for free stuff to help with your health and fitness.
See ya!
Are you …
Smart
Driven
Creative
Optimistic
Hardworking
Forward thinking
But also …
Tired
Isolated
Frustrated
Overwhelmed
Not sure where to start
I’m Building a Network
This is a brand new and growing initiative to bring together people who choose to make things happen.
We need smart hard-working people to build the future