Anthropic Is Having an Incredible Run. I Just Don't Think It Will Stay on Top for Long.

There is a lot of hype around Anthropic and its potential IPO right now, and honestly, it's easy to see why. The numbers are staggering. The company reported a revenue run rate of more than $47 billion in May, Reuters says that figure has already climbed past $65 billion as of the end of July, and projections of $190 to $200 billion in revenue by 2028 are fueling talk of a valuation approaching $2 trillion.

That's incredible. But here's where I struggle with the valuation: what does being the leader in AI today actually mean three years from now?

I've been building technology and digital businesses for more than 25 years. I've watched the internet, search, social media, and mobile completely change how companies operate, and I have never seen anything move at the speed AI is moving. That's exactly what makes stretching Anthropic's current trajectory out several years so difficult for me.

Don't get me wrong. Claude is excellent, and Claude Code has been a genuine hit — Anthropic said earlier this year it was approaching a $2.5 billion run rate on its own, with enterprises driving more than half of that. There is clearly a tremendous business here. But believing Anthropic will keep being successful and believing it will keep dominating are two very different things, and I don't think enough people separate them.

Today's AI advantage can vanish fast

Think about how quickly this industry turns over. One model drops and suddenly everyone is talking about it. A few months later, somebody ships something better. Capabilities that looked remarkable six months ago become expected features. Prices fall, models get faster, open alternatives improve, new companies show up, and the tech giants keep folding AI into products that already have enormous distribution. This market isn't moving annually — it's moving monthly, and sometimes weekly.

That's why I can't look at today's growth curve and simply draw the line out to 2028. Anthropic is competing against some of the best-funded, most capable technology organizations in the world, and the risks — brutal competition, massive infrastructure costs, falling AI prices — are very real. And there's another risk that's even harder to put a number on: the company that eventually disrupts Anthropic may not be a company any of us are talking about today.

AI may not crown a single winner

We tend to judge every new technology by the ones that came before it. Google owned search, Facebook owned social, Amazon owned e-commerce, Microsoft owned the operating system — so naturally everyone is hunting for the equivalent winner in AI. I'm not convinced this market develops that way.

Businesses are already comfortable running multiple AI platforms side by side. Claude for coding. ChatGPT for another workflow. Gemini because they live in Google's ecosystem. A specialized model for their industry, and something that doesn't exist yet for whatever comes next. If the gap between the leading models keeps narrowing, the real advantage shifts toward price, distribution, integrations, and workflow rather than who has the smartest model. That makes sustained dominance very hard to hold onto.

$65 billion is incredible. $200 billion is something else entirely.

Anthropic deserves enormous credit for what it has pulled off — a run rate that went from roughly $9 billion at the end of 2025 to more than $47 billion in May, and now reportedly past $65 billion. But the valuation thesis asks investors to believe the company can approach $200 billion in annual revenue by 2028 while holding a multiple reserved for the world's most strategically important technology companies. That's a huge bet. Not because Anthropic isn't a great company, but because AI doesn't stand still long enough for me to comfortably assume today's competitive advantages will still be advantages three years from now.

I wouldn't bet against Anthropic. I'd bet against permanence — and that's an important distinction. I think Anthropic could become an enormous public company. I think Claude will remain an important platform, and I think revenue could grow substantially from here. But if you asked me whether Anthropic will still be the top dog in AI three years from now, I wouldn't take that bet. Not because I know who beats them. That's precisely the point. Three years in AI feels like fifteen years in almost any other industry. The disruptor might be a company nobody has heard of yet, a breakthrough that makes today's models dramatically cheaper to run, a specialized tool that businesses simply prefer, or a major leap from Google, OpenAI, or Meta. Or Anthropic could keep leading. Nobody really knows.

Anthropic's biggest challenge may not be getting bigger. It's staying ahead while everything around it is moving just as fast — and there is no historical playbook for that. Anthropic is having an incredible run. I just wouldn't assume today's leader will still be tomorrow's leader. And when valuations start approaching $2 trillion, that's a pretty important distinction.