The Internet Is Going to Need More Names

For a long time, the internet trained us to believe that websites were old-fashioned.

The feed won. The platform won. The algorithm won.

Most people stopped thinking of the internet as a place you visited and started thinking of it as a stream you refreshed. Facebook, Instagram, TikTok, LinkedIn, YouTube, X — they all taught the same behavior: don't go somewhere, just keep scrolling. Your identity became a profile. Your thoughts became posts. Your memories became content. Your audience became whatever the algorithm decided to let through that day.

Then AI showed up and quietly changed the math.

Not all at once. Not in the obvious way. The first-order reaction was about content: AI can write, summarize, design, code, remix, imitate, answer, and generate. That part got everyone's attention. But the deeper shift may not be about content creation at all.

The deeper shift may be about naming.

Because when content becomes infinite, the scarce thing is not another paragraph, image, video, landing page, or chatbot. The scarce thing is the human decision to claim something, name it, locate it, preserve it, and say: this is mine, this came from me, this matters enough to have a place.

That is why I think domain names are about to become more important, not less.

Not because every domain is valuable. Not because domain speculation is automatically smart. Not because adding "AI" to a name turns it into gold. That is the shallow version of the story.

The better story is this:

AI collapses the cost of creating websites. But it increases the value of meaningful addresses.

For years, the hardest part of having a personal website was building the thing. You needed design skills, development skills, writing skills, hosting knowledge, or enough money to pay someone else. Most people didn't have that. So the web concentrated inside platforms. We posted where it was easiest to post.

AI and vibe coding are changing that. Suddenly, one person can create a microsite, an essay, a visual system, an interactive experience, a narrated article, a PDF, a landing page, a searchable archive, or a weird little internet object in a weekend. Sometimes in a night.

That changes the role of the domain.

The domain is no longer just the thing you buy after you already have a business. It becomes the first act of authorship. A domain is a public claim. A name. A container. A timestamped decision. A little piece of language pulled out of the infinite stream and given an address.

The numbers already point in that direction.

At the end of the first quarter of 2026, the domain-name system had 392.5 million domain registrations across all top-level domains, up 24.1 million year over year. That is not all because of AI, but it is the backdrop: the naming layer of the internet is still growing, not fading away. Source: DNIB.

Then look at what happened specifically around AI. DNIB, the Domain Name Industry Brief, traced a clear inflection point after ChatGPT's public release in November 2022. New generic top-level domain registrations ending in "AI" — names like ExampleAI.net — totaled 300,000 names in 2023, growing 67% year over year. DNIB also reported that the .ai country-code domain base increased 142% year over year to 359,000 names by the end of 2023. Source: DNIB.

By the end of 2025, .ai had reportedly surpassed one million registrations. Hogan Lovells described the growth path clearly: .ai went from roughly 107,000 registrations in 2020, to 148,000 in 2022, to 359,000 in 2023, to 800,000 in 2025, before crossing the million-name mark. Source: Hogan Lovells.

That is not just a domain trend. That is culture moving into infrastructure.

A tiny country-code extension for Anguilla became a global semantic label for artificial intelligence. The letters "AI" moved from a technical abbreviation into everyday language, then into startup branding, then into domain registrations, then into national revenue. Anguilla's domain extension became part of the global AI economy because meaning changed faster than geography.

That is one of the most interesting things about domains: they are technical objects, but they behave like language.

.tv became media. .io became tech. .me became personal identity. .ai became artificial intelligence. The original administrative meaning remains, but the cultural meaning takes over.

This is where the domain market becomes more than a market. It becomes a map of what people believe the future is going to need.

And right now, people are registering names as if the future is going to need a lot of AI-shaped identity.

Namecheap's 2025 domain report adds another piece. It says .com continued to dominate, but alternative TLDs kept gaining broader acceptance. In Namecheap's own registration mix, .com represented 40.1% of new registrations in 2025, while non-.com domains represented 59.9%. The report also says .ai and .app continued climbing, and three of the five highest-priced Namecheap marketplace sales in 2025 were .ai names, led by blockchain.ai at $405,000. Source: Namecheap.

That does not mean every .ai name is worth money. It means the market is telling us something: when a new technological category becomes culturally obvious, the naming layer gets crowded fast.

But the most important argument is not about .ai.

The most important argument is about what AI does to the structure of the internet.

A recent domain-landscape analysis by David Barnett and Lars Jensen argues that AI could drive a "fundamental step-change" in domain demand. Their point is not just that more people will register AI-related names. Their bigger point is that AI agents, machine-to-machine systems, connected devices, automated web activity, and new AI-enabled businesses may all require more persistent, trusted identifiers. In that world, a domain can function almost like a "birth certificate" or trusted identity root for an AI system.

That phrase — birth certificate — is the part that matters.

Because if AI agents become common, and if those agents act on behalf of people, companies, institutions, workflows, products, and communities, then the question becomes: who is this agent, where does it live, who controls it, and why should I trust it?

A domain can help answer that.

Not perfectly. Not alone. DNS is not magic. A domain does not prove virtue, quality, honesty, authorship, or safety by itself. Bad actors can register domains too. But domains are still one of the internet's oldest human-readable trust systems. They are memorable. They are resolvable. They can be verified. They can host records. They can point to content. They can persist beyond the platform.

And maybe most importantly, they force a small act of friction.

You have to choose the name. You have to register it. You have to pay for it. You have to renew it. You have to decide what belongs there. You have to connect it to something.

In a world of infinite AI output, friction becomes signal.

That may sound backwards because we usually talk about technology as a force that removes friction. Faster is better. Cheaper is better. Easier is better. One-click is better. Instant is better.

But meaning does not work that way.

Meaning often requires resistance. Not pointless resistance. Not bureaucracy. But enough resistance to show intent.

Anyone can prompt an AI model to generate 100 names. Anyone can ask for 50 logos. Anyone can produce a landing page. Anyone can flood the internet with plausible content. But not everyone will claim a name, build a place, preserve the origin, connect the evidence, maintain the address, and keep returning to it.

That is the difference between content and authorship.

AI makes content abundant. Domains can make authorship locatable.

This is why I think the next version of the internet may be less about "websites" in the old sense and more about places.

A place is different from a page.

A page is something you land on. A place is something you return to.

A page can be generated. A place has accumulated meaning.

A page answers a query. A place carries a point of view.

A page can be summarized. A place has texture, memory, sequence, and intent.

That is the piece most AI-search conversations miss. People talk as if AI summaries will replace websites because users won't need to click anymore. In some cases, that will be true. If the question is simple — what time does the store close, what is the weather, how many ounces in a cup — the AI answer may be enough.

But not every human need is a query.

Sometimes we are not looking for an answer. We are looking for a person. A worldview. A body of work. A trail of thinking. A trusted source. A signal that someone has been wrestling with the same thing we are wrestling with.

That is where personal websites become interesting again.

Not because they can out-distribute social platforms. They can't.

Not because they can out-answer AI. They won't.

But because they can do something neither the feed nor the chatbot does well: they can hold together a person's intent over time.

That is what I want Ryans.art to become. Not a feed. A place.

A place for AI-amplified thinking. A place where an idea can start messy, become structured, branch into essays, visuals, audio, microsites, PDFs, thought pieces, and experiments. A place where the work is not trapped inside the logic of one platform's engagement machine. A place where the URL itself says something: this is not just content; this is a human trying to make meaning with new tools.

That is also why .art matters to me.

I don't think .art is only for paintings, galleries, or traditional creative work. I think .art is becoming a useful label for human expression in an age when the definition of art is expanding again.

Photography changed art. Recording technology changed music. Film changed storytelling. The internet changed publishing. AI is changing expression.

The question is not whether AI-generated work is "real art" in some simplistic yes-or-no way. The better question is: where is the human intent?

Who chose the frame? Who asked the question? Who preserved the output? Who connected the dots? Who decided this mattered? Who gave it a name?

That is why a domain can be more than an address. It can be a container for intent.

This is also where domain names connect to provenance.

For most of the internet's history, we treated provenance as a problem for museums, journalism, academia, finance, or legal records. But AI is going to make provenance a daily problem. When content can be generated endlessly, origin matters more. When images can be fabricated, timestamps matter more. When voices can be cloned, source trails matter more. When search becomes summary, the question of what gets cited, remembered, and trusted becomes more important.

Domains are not the whole answer. But they are part of the answer.

A domain can hold DNS records. It can point to a public artifact. It can connect to hashes, timestamps, archives, authorship statements, wallets, registries, source files, essays, and version histories. It can become a root for a proof system that says: this was claimed here, by this person or entity, at this time, with this evidence.

That is the deeper version of the HUMAiN.art thesis: not domains as collectibles, but domains as proof-of-intent containers.

The speculative version of this idea is huge. Barnett and Jensen even float the possibility that the domain landscape could grow from hundreds of millions of names today to tens of billions by 2050 if agentic AI, connected devices, machine-to-machine systems, and identity infrastructure evolve in that direction. That is not a forecast anyone should treat as guaranteed. But it is a useful thought experiment because it forces the right question: what happens if the internet needs names not just for companies and websites, but for agents, tools, workflows, artifacts, identities, and human-authored origin points?

The conservative version is still meaningful.

Even if we never reach tens of billions of registered domains, AI is already changing naming behavior. People are registering AI names. Companies are creating AI subdomains. Investors are buying AI-related digital real estate. Startups are using alternative TLDs to signal category, function, and identity. Creators are discovering they can build owned media properties without needing a full development team.

That is enough to matter.

The mistake would be to reduce all of this to a gold rush.

Gold rushes happen. Bubbles happen. Bad names get registered. People overpay. Trends fade. Renewal bills arrive. Most domains will not become valuable. A name by itself is not a business, an audience, a product, a movement, or a work of art.

But bubbles often form around real shifts.

The dot-com bubble did not mean the internet was fake. The crypto bubble did not mean digital ownership questions disappeared. The AI hype cycle does not mean AI is irrelevant.

Sometimes speculation is the chaotic shadow cast by a real structural change.

The structural change here is simple:

Creation is getting cheaper. Trust is getting more expensive.

That one sentence explains a lot.

It explains why brands still care about .com.

It explains why .ai exploded.

It explains why alternative TLDs are becoming more acceptable when they communicate something specific.

It explains why personal websites may return.

It explains why provenance matters.

It explains why a domain can be both a boring technical record and a deeply human object.

The boring version of a domain is that it maps a name to an IP address.

The human version is that it maps intention to memory.

That is the part I can't stop thinking about.

A domain is one of the few remaining internet objects that still feels like claiming a place. Social posts are temporary. Feeds move on. Platforms redesign themselves. Algorithms change. Accounts get throttled, hacked, suspended, abandoned, or buried. But a domain has a different feeling. It is closer to putting a stake in the ground.

This does not mean everyone needs hundreds of domains. Most people do not.

But I do think more people will need one meaningful domain.

A personal domain. A project domain. A family archive. A company idea lab. A public notebook. A proof-of-work trail. A portfolio of experiments. A place where their AI-amplified thinking can live outside the feed.

In the old internet, the website was the product.

In the platform internet, the profile became the product.

In the AI internet, the domain may become the root.

The root of identity. The root of trust. The root of authorship. The root of a human-readable place in a machine-readable world.

That is why this moment feels bigger than .ai.

.ai is the obvious headline because it has numbers, sales, hype, and a clean story. But the more interesting question is what happens after everyone has AI.

When AI is no longer a differentiator, what becomes valuable?

Taste.

Trust.

Memory.

Point of view.

Human context.

Original naming.

Proof of intent.

The ability to create a place that feels like it could only have come from one person, one team, one community, or one moment in time.

That is where domains matter.

Not every domain. Not every trend. Not every registration. But the right name, attached to the right human intent, preserved in the right way, can become a durable container for meaning.

The internet is not running out of content.

It is running out of trusted places to put it.

AI will generate more words than anyone can read, more images than anyone can see, more summaries than anyone can verify, and more answers than anyone can trace. The future problem will not be whether something can be created. The future problem will be whether something can be located, trusted, remembered, and connected back to a human source.

That is why I keep coming back to the domain.

Small. Old. Technical. Unsexy.

Still powerful.

A domain is a name with an address.

A name with a record.

A name with friction.

A name with memory.

And in a world where machines can generate almost anything, the human act of naming something may become one of the most important creative acts left.