Here’s a fun fact: the guy who literally invented the internet is now watching AI companies make the same mistakes he watched the early web avoid. Vinton Cerf, the 83-year-old networking legend, just dropped some wisdom at the Open Frontiers conference that should make every AI investor sit up and pay attention.
Cerf’s core message? AI is heading toward the same inflection point the internet hit decades ago—and if the industry doesn’t learn from history, it’s going to repeat it.
The Walled Garden Problem
Remember when the internet could’ve been owned by one company? Yeah, that didn’t happen, and it’s the entire reason you’re reading this on the open web instead of AOL’s proprietary network. Cerf’s point: the internet only became ubiquitous because nobody owned it. Everyone could plug in, follow the same protocols, and suddenly a university in California could talk to a government lab, which could talk to a commercial ISP. Same language, same rules, infinite possibilities.
Now look at AI. We’ve got OpenAI’s ChatGPT, Google’s Gemini, Anthropic’s Claude—all these closed ecosystems that don’t really talk to each other. Cerf’s warning is blunt: as AI agents multiply and start coordinating with each other, we’re going to need interoperability and standardization. Otherwise, we’re building a fragmented mess instead of a platform.
The Language Problem (And It’s Not What You Think)
Here’s where it gets interesting. Cerf says natural language—English, Mandarin, whatever—won’t cut it for AI agents communicating with each other. Why? Because human language is ambiguous. We rely on context, inference, and the fact that “bank” can mean a financial institution or the side of a river, and we figure it out.
Machines can’t do that. When one AI agent agrees to do something with another AI agent, both need to be absolutely certain they understood the same thing. Cerf put it perfectly: “An agent really needs to be sure the other agent understands what it is that they just agreed to do together.” That’s not poetry—that’s engineering. And it matters.
Platforms Beat Products
The final lesson is almost obvious once you hear it: the biggest technologies don’t win by being standalone products. They win by becoming platforms that let other people build on top of them.
Google, Amazon, Netflix—none of these companies built the internet. They built on it. The internet was the enabling technology. Cerf’s saying AI needs to follow the same playbook. The real winners won’t be the companies with the fanciest chatbots. They’ll be the ones who create the foundation that lets thousands of developers build the next generation of applications.
Why This Matters for Your Portfolio
If you’re investing in AI, this is the framework to watch. Companies obsessed with proprietary moats and closed systems? They’re fighting the tide. The ones building open standards, enabling interoperability, and creating platforms? That’s where the long-term value lives.
Cerf didn’t invent the internet to watch it get rebuilt as a bunch of walled gardens. He’s not about to let AI make the same mistake.
