Vitalik Buterin’s recent update to Ethereum’s roadmap, inspired me to write this article. The community’s response has been mixed, with debates spanning developer retraining, the complexity of the tech, and the long-standing requirement of EVM compatibility for new chains and dApps. If you’re unfamiliar with the details, if ZK, EVM or RISC-V doesn't ring a bell - no worries. It’s not essential, and delving into technicalities might derail our focus. With that in mind, I’ll shift to what’s emerging from Ethereum’s foundation and why I’m genuinely excited about this direction.
Vitalik is officially laying the groundwork for Web4. In my view, it’s long overdue.
My journey into this field began in 2015 with my first IoT project: a system of smart scales for warehouse inventory tracking. In 2016, I entered the blockchain ecosystem, contributing to a fork of IOTA, which was briefly celebrated as a blockchain 2.0 solution for the Internet of Things. Later, as I delved into data analytics, I discovered a critical third component. These experiences convinced me that the future of the Internet lies in the synergy of IoT, distributed ledgers, and artificial intelligence. Web4 will mark their culmination - but for whom the bell tolls? Let’s explore.
Picture a cosy evening in an old data centre, the hum of servers in the background, where two analysts, like characters straight out of a cyberpunk novel, are arguing about the internet’s future. Their conversation isn’t just idle chatter - it’s an attempt to peer into a world where Web4 is already reshaping reality. This isn’t just tech; it’s a philosophical dilemma: who are we building this digital future for - humans or machines? In this article, I’ll take you from the internet’s origins to the horizon of Web4, leaning on my experience in IoT, big data, analytics, and blockchain to show how tech is shaping a new world - and whether we can stay in charge.
As usual, in the style of rational realism-optimism :)
After a long preamble…
Analyst 1: Here’s the central question. Understand it and explain it to others.
Analyst 2: Understand what?
Analyst 1: Who are Web3, Web4, and beyond being built for? Humans or machines?
Analyst 2: That’s unexpected… Likely for machines, I suppose.
Analyst 1: Elaborate. What led you to that conclusion?
Analyst 2: Two factors: First, Web3 and Web4 will integrate global AI with machine control, marking a revolutionary leap from Web2 for machines. Second, for humans, Web3 and Web4 represent incremental progress, not a transformation. We’ve nearly reached the limit of network interaction. The future lies with Androids and AI. Thus, the evolution of the internet and AI is more critical for machines than humans.
Analyst 1: That's right!
To get what our analysts are debating, let’s take a whirlwind tour of the internet’s history—from its birth to the edge of Web4. There’s a ton of info out there, so I’ll keep it high-level. I won’t dive into my crypto-anarchist grandpas chasing their ideals from manifestos to the Principality of Sealand. Nor will I talk about today’s crypto-kids, chilling in Miami, Dubai, or Singapore with their Lambos, free of all that broke-ass nonsense. Nope. A quick rundown of the key points, and we’re off to the main topic.
Web1 - Read (1990s–early 2000s): The pioneer internet. Static HTML pages where you could only read. Yahoo!, AltaVista, and early browsers like Netscape were your windows to the digital world. It was like a library with no interactivity—you grabbed a book but couldn’t scribble in the margins. The magic!
Web2 - Write (2000s–2010s): The internet went “read-write.” Social media, blogs, YouTube, and giants like Google and Facebook turned users into content creators, not just consumers. But freedom came at a cost - centralisation. Your data belonged to corporations, not you. We ended up in the era of “Likes” and digital feudalism.
What it was: the internet stopped being a library (Web1) and became a platform for user-generated content.
Key players: Facebook, Google, YouTube (yeah, also Google) - they gave people a voice but took their data.
Business model: “Free cheese” (you’re the product, ads are the cash). “If you’re not paying for the product, you are the product.”
How it looked 10 years ago: “Hell yeah, anyone can be a blogger!” “Social media will unite the world!”
What we got: censorship, algorithmic echo chambers, digital slavery. User data became the new oil, and users - free oil rigs.
Web3 - Own (2010s–present):
A pushback against Web2’s monopoly. Blockchains, smart contracts, and decentralised apps (dApps) gave control back to users. Ethereum, Solana, and NFTs became symbols of an era where you don’t just read and write - you own a piece of the digital world. In theory…
What they promised: decentralisation! No more Zuckerbergs - just blockchain and crypto. Own your data! NFTs, DeFi, DAOs. A new economy! Tokens over Likes.
How it looked 5 years ago: “Banks are toast!” “Art will belong to artists!”
What happened: instead of liberation, we got a speculative circus (NFT monkeys, scams, gas wars). Blockchain turned out slow and pricey, and decentralisation? Well, most nodes are run by cloud providers.
Irony: the same funds ruling Web2 scooped up crypto.
Reality in 2025: Web3 didn’t kill Web2 - it became its crypto sidekick. Meta’s dropping NFTs, Google’s using blockchain for cloud, and Twitter (X) is rolling out DOGE payments. The rebellion flopped, but the tech stuck around.
A quick aside… I’m all in on the early cypherpunk vibe and generally root for crypto. Hell, my team and I built a few DeFi protocols, working on Tezos, Ethereum, Bitcoin, and Polkadot, even snagging some grants. I believe in tokenisation and think security tokens are the only elegant fix for the wild money-printing press… but the modern crypto-gypsy scene? It gives me pain.
Web4 (future): This isn’t just the internet anymore - it’s a fusion of the physical and digital worlds. The Internet of Things (IoT), artificial intelligence, and distributed ledgers merge into a network where machines talk to each other, and humans are part of a smart ecosystem. Web4 promises automation, security, and autonomy - but at whose expense? Will your coffee maker mine crypto while Amazon still skims a cut?
Back when IoT was peak hype, its Achilles’ heel was device security. The Mirai botnet may be “defeated” now, but it wasn’t the last to effortlessly knock out droves of “smart” devices. Today, modern protocols like zero-knowledge (zk-SNARKs, zk-STARKs) are tackling this issue. But the big question remains: who’s all this for?
Analyst 1: Next question: who’s better for humans to deal with “on business”? Forget drinking, saunas, or playing pool. For practical stuff, who’s the better partner: a machine or another human?
Analyst 2: That question’s off-base.
Analyst 1: Why?
Analyst 2: How do you compare them? What’s the metric?
Analyst 1: Picture machines acting like humans. Take a gun, a plane, a car, a computer, a phone. How’d they behave?
Analyst 2: A gun would misfire 7 out of 10 times and shoot you in the face once. A plane would land on some random airstrip… if you’re lucky. A car wouldn’t start, or if it did, it’d refuse to speed up at a critical moment and say, “We need to talk.” A computer would demand alimony, hog two-thirds of the screen, and eat four-fifths of your RAM. A phone would leak your private chats online and snitch your coordinates. Why? Because of betrayal, inflated egos, corruption, whims, emotions, “I’m out” drama - that’s all human crap.
Analyst 1: Spot on. Now consider: humans can use machines without guilt, send them to their doom, and owe them nothing—not to them or their “kin.” So, humans prefer machines over other humans. Humans oppress, betray, and compete with each other, but machines serve loyally. Web3 and Web4 are building a world - or, if you like, a Noosphere - for machines, right?
Analyst 2: Seems that way… but that’s terrifying!
Analyst 1: Why terrifying?
Analyst 2: Because we’ll become appendages! Humans won’t be needed!
Analyst 1: That’s cute. Some humans, if progress keeps going, will upgrade their bodies and brains, keeping just their consciousness. That’s basically a self-aware machine - a real prospect, and not far off.
Analyst 2: Who’d allow that?
Analyst 1: Allow? Humans are building it with glee.
Analyst 2: What do we do? I can’t picture that world…
Analyst 1: Two paths: smash progress, wreck everything, throw humanity back, or charge forward, but curb elitism and craft humane principles for a society with AI and machines. Those principles will show up like crypto did.
Analyst 2: Why not now?
Analyst 1: Because humans are stuck thinking in pyramids. Elitism’s baked into our bones. Dictators, leaders, and their cronies—it’s like the 21st century is still the Middle Ages. Nationalism’s part of that baggage.
“In Web4, you’re either a prompt engineer or a digital peasant.”
This dialogue isn’t just philosophical banter - it’s a mirror reflecting tech’s trajectory. Web4 will fuse IoT, blockchains, and AI. Smart fridges will order groceries, cars will swap traffic data, and sensors will track everything from your pulse to CO₂ levels. But IoT’s got a weak spot - security. Modern zk-protocols, like zk-SNARKs, fix this, ensuring privacy and data protection. It’s like a digital safe: nobody knows what’s inside, but everyone trusts it’s secure.
But who’s this all for? The analysts hint: that machines are winning. They don’t betray, whine, or demand raises. Web4 makes them autonomous, smart, and almost alive. And humans? We risk becoming “digital peasants” servicing machines or “prompt engineers” feeding questions to AI.
I know several programmers who are reading my blog, so here is something for you, lads. If you are not into coding, this part of the text will be looking at the same problem with very similar aspects but from a niche angle. So if you are not - you don't have to, as it will be repeating what we have discussed already :) There are a lot of jokes popular among architects and senior-level coders regarding AI and coding. These jokes are usually about as amusing as watching paint dry.
Example:
— AI will be able to create complex codebases from scratch // Dario Amodei, CEO of Anthropic
— We’d be properly worried if it could cobble together simple code by refactoring, but this? Complex and from scratch? Anyone can do that.To grasp how Web4 will reshape the world, let’s look at programming - because, to me, programming as we know it is a key to understanding Web4. There are two core principles in coding or learning to code:
Code Readability: From day one, coders are taught to write code others can understand. But if AI starts writing software for machines, readability won’t matter. Why would a human need to parse code optimised for GPUs?
Code Efficiency: Languages like C are fast because they’re built for the architecture of ancient processors. But AI thrives on graphic accelerators, like NVIDIA’s. Linus Torvalds, a C diehard, trashes AMD, but the future belongs to architectures tailored for AI. C will fade, and so will old paradigms.
Web4 could spawn software where readability or speed takes a backseat to AI compatibility. Programmers will become conductors, not code writers. It’s daunting but opens doors. Already, any expert who’s legit in their field and uses AI tools outpaces those who don’t. But there are hurdles: crafting effective prompts takes time, and checking for inaccuracies or AI hallucinations is a slog. Still, these are temporary limits - AI tools are evolving fast.
Web4 isn’t a dystopia - it’s a challenge. Zk-protocols will secure devices, and blockchains like Ethereum will ensure trust. But the real question is: how do we, humans, stay in control of this world? The analysts’ chat suggests: without new principles, without ditching elitism and power pyramids, we’ll end up as machine appendages.
There are two paths: smash progress or charge forward, but smartly. Web4 can be a world where machines serve humans, not the other way around. My goal is to help build those solutions, working with funds and startups that share this vision. It’s not a fantasy - it’s a task. And we need to tackle it now.
