Public Good or Public Hype?
Tan argues that frontier models themselves trained on public human knowledge so access to capable AI should be “a form of public good.”
Rebranding the Same Idea
It’s not the first time someone has claimed AI should be a public good. The idea has been around for years, but now it’s dressed up in new jargon and called a “frontier model.”
What’s different this time? Nothing. The same old argument, just with more buzzwords and less accountability.
Tan is right that AI models are built on public knowledge, but that doesn’t mean they should be free. It just means they’re built on the same old data we’ve all been using for years.
The real question isn’t whether AI should be a public good. It’s whether companies will ever stop pretending they’re doing something new when they’re just rehashing the same ideas.
The Same Old Song
It’s easy to say AI should be a public good when you’re not the one funding the models. The reality is, training these systems costs millions, and no one wants to pay for it.
Tan’s argument is valid, but it’s also not new. It’s the same argument that’s been made about the internet, open-source software, and every other technology that’s supposed to be a “public good.”
What’s missing is a plan. No one is talking about how to make AI truly accessible without breaking the companies that build it.
It’s not that the idea is bad. It’s that it’s being sold as a breakthrough when it’s just a rebrand of an old concept.
The real innovation isn’t in the idea. It’s in the way it’s being presented — as if it’s something new, when it’s just the same old thing with a different label.
Public Good or Public Theater?
Public good is a nice-sounding phrase, but it’s also vague. What does it mean? Who decides what’s a public good? And who pays for it?
These are the questions that Tan and others are avoiding. Instead, they’re focusing on the idea that AI should be free, without addressing the practicalities.
It’s not that the idea is wrong. It’s that it’s being used as a marketing tool, not a real solution.
When someone says AI should be a public good, they’re not just talking about access. They’re talking about control, responsibility, and who gets to decide what’s allowed.
And that’s the real issue. Not the idea itself, but who gets to shape it and who gets to pay for it.
Same Idea, New Name
Calling something a “public good” doesn’t make it one. It just makes it sound more virtuous.
Tan’s argument is not new. It’s just being presented with more flair and less substance.
The problem isn’t the idea. It’s the way it’s being sold — as if it’s a revolutionary concept when it’s just a rehash of an old one.
Public good is a term that’s been used for decades. Now it’s being applied to AI, and that’s what’s getting people excited.
But excitement doesn’t change the fact that the same old ideas are still being sold as something new.
