It’s no secret that the death of the internet as we know it is no longer greatly exaggerated. Between the big tech giants and generative AI the death spiral of the web has accelerated, like some digital ouroboros, the internet has begun to feed on itself and shit out the worse parts for the rest of us to consume. Web3 failed to deliver anything but more similar in the form of cryptocurrencies, NFTs and ultimately a market-fuelled distraction cobbled together on top of blockchains which is hardly that venture capitalists are determined to push as the correct rhetoric — one that is self-serving because of their investment into get-rich-quick meme coins rather than the projects themselves. If we look back at the web it began with one man’s PC sitting on his desk with a post-it note warning people not to turn it off. It then took 2 years of fucking around with DNS, TCP/IP and HTML before the first 50 websites emerged and then things took off nicely; it was under control of no one and it was decentralised. enshittification the new internet Sadly it wasn’t to last. What began as something owned by everyone on the planet became something owned by few and rented out to everyone for a price, that invoice was either paid financially or through data which the latter we are now beginning to understand the full extent of because it’s being put to good use by the generative AI startups with no compensation. And that’s coming back to bite back hard so very quickly. It’s poetic that Facebook will die choking on its own vomit soon. AI-derived content has taken over the internet and it’s unstoppable and almost unrecognisable as something an algorithm generated to the untrained eye. Those in the know are a privileged few, the vast majority of people wouldn't know where to look to spot a faked article or image, recognise telltale signs, or even copy into an AI detector to confirm its source and origin. The long term consequences are such that pretty much nothing on the internet will be human created or curated, nor even read and digested by people either. We’re being pushed to adopt personal AI assistants to consume the content they gave birth to then regurgitate it back to us as a synopsis for us to use elsewhere or even take action on either personally or in a business context. This is the antithesis of what Tim Berners-Lee wanted to see out of the Semantic Web (or Web 3.0) back in the late 90's— an internet where users kept their data sovereignty, instructed AI agents to perform tasks for them but the web itself was very much human. Web3 startups claim that they have the answer, that blockchain is somehow a magical silver bullet for data provenance and to train new types of AI on, and another excuse for another round of meme coins and but this house of cards is rapidly falling down. Like a stack of dominoes one by one very public and in your face Web3 projects like Nike’s RTFKT and are quietly being abandoned, with claims that intelligent NFTs Starbucks’s Odyssey it was a proof of concept and we’ll do better next time. Only there won’t be a next time for them. They overestimated how interested their existing customer base would be, underestimated how much the hype would be driven by the same circle of Discord lurkers shouting WAGMI and wen moon, and grossly underestimated and understood just how apathetic the general public and those who just want to drink a bloody cup of coffee would be to collecting for a glorified Pannini sticker album. What I will say is that blockchain isn’t the answer, it never was. It’s a distributed ledger you can do stuff with but it’s not an internet protocol or replacement for the web itself. The fact that you need to create and burn magic internet money to make it work invites speculative behaviour. Web3 turned the internet into a different type of stock market, trading made-up asset classes and foreign currencies and completely abandoning why it existed in the first place. After the death of the first internet, lots of tiny new internets will grow up in its place. Philip Rosedale, creator of Second Life There’s an interesting turn of events emerging though, a strange confluence of ideas mixed with a societal need for nostalgia. We’re seeing a shift across generations that is railing against the relentless push towards subscriptions and digital ownership People are buying physical media in all forms, old and current formats are enjoying a revival as subscriptions become ever more expensive, overlap each other and the choice that we thought was a convenience is now strangling us. because the reality is that you own nothing under either model. The digital feudal system has robbed us all of what we held dear before and web3 failed to give us control back. The internet and how it runs now is dead or drawing its last breath. But it won’t be replaced by one single, new internet nor will it be replaced by blockchains. It’ll be replaced by networks. We have some examples already like and but these rely on blockchains and ultimately are no good both longer term but also from a decentralised control perspective. Another example is , a small group that’s been working for nearly two decades on building a distributed network protocol that works with ordinary kit you’d find in the home (recently announced rebrand as now) IPFS Arweave SAFE Autonomi Much like the distributed systems that allowed BitTorrent (distributed file sharing) and BOINC (distributed computing like SETI@Home and Fold@Home) to function on consumer-grade hardware, there will be a re-emergence of the same movement where people will power the next iteration of the internet from their own devices — not from centralised cloud stacks owned by the same 3–4 major corporations. The internet itself was build as a decentralised and distributed network before capitalism found a way to make it work to its advantages. You can see how this could come about after recently watching the NVIDIA GTC keynote and understanding Sam Altman’s own drive to create a new silicon empire — they’re scared there is not enough compute power that they can build and control for their own ambitions, and control is the key point to pay attention to here because there is already enough compute and storage out there sitting on billions of devices in the world but they just won’t control any of it. When robotics start taking central stage over and above incorporeal AI systems then you’ll begin to understand that every robot in itself is a distributed node on a new type of compute network that won’t rely on centralisation for decision making, but as a collective hive mind can improve and self-learn at an exponential rate. But this is an example of a distributed network under centralised control. And it doesn’t stop at being just one internet, the revolution will be that there will be many under this model, all owned by nobody but powered by everyone in those particular communities it supports. What we need for the internet to be remade is not only a new set of distributed networks and designed to operate in a decentralised manner but also that the network, storage and compute are owned by nobody but powered by everybody. operating systems It was a reason I believed that Web3 DAOs (decentralised autonomous organisations) could have taken back control from local authorities as community driven projects, wholly owned and run by the people in those local communities, empowering them to make decisions that made a difference rather than line an MPs pockets. When folks like Peter Diamandis claim that AI will usher in an Age of Abundance what they’re really calling for is a rewriting of economic systems beyond the capitalist one we use today, and that starts by rewriting the infrastructure we rely on so much to power it — the internet. The internet is dead. Long live networks. Also appears . here

Too Long; Didn't Read

Discover Saakuru: Web3 Mass-Adoption with 0-FEES

The Internet Is Dead. Long Live The Internet.

Too Long; Didn't Read

Theo Priestley

About Author

TOPICS

Languages

THIS ARTICLE WAS FEATURED IN...

The Internet Is Dead. Long Live The Internet.

Too Long; Didn't Read

Theo Priestley

About Author

TOPICS

Languages

THIS ARTICLE WAS FEATURED IN...

RELATED STORIES