Burning the Library

Robert Thomson’s ruthless "Woo and Sue" playbook isn't just blocking AI; it’s turning the open internet into a gated community.

[Speaker 1]: It happened this morning. Another wall went up. Meta just announced they’ve signed a fifty-million-dollar-a-year deal with News Corp. That means the content from the Wall Street Journal, the New York Post, and the Times of London is now officially flowing into Meta’s AI models-and pretty much only their models. [Speaker 2]: Right. And this follows the pattern we’ve seen all year. First it was Reddit and Google, then OpenAI and Axel Springer. Now Meta. The era of the "open web"-where you could just scrape the internet and build a smart tool-is effectively over. It’s been replaced by what analysts are calling the "partitioned web." [Speaker 1]: And on the surface, this sounds like a boring copyright dispute. Big companies trading checks. But there is a massive side effect happening right now, in real-time, that affects anyone who uses the internet. Because to protect that fifty-million-dollar asset, these publishers aren't just locking the doors. They’re burning the library. [Speaker 2]: Exactly. They are blocking everything. Not just the AI scrapers, but the archivists, the researchers, and the historians. [Speaker 1]: We’re looking at why the internet has quietly transformed from an open library into a gated community. And we need to talk about that number from the BBC study last fall-the forty-five percent error rate. Because if this trend continues, your digital assistant isn’t just going to get more expensive. It’s going to get significantly dumber. [Speaker 2]: It’s Wednesday, March 4, 2026, and you’re listening to The Angle. [Speaker 1]: So, to understand why this Meta deal matters, we have to look at the strategy that got us here. Because this didn't happen by accident. This was a very specific playbook designed by the legacy media companies. [Speaker 2]: Yeah. It’s the "Woo and Sue" strategy. [Speaker 1]: Right. "Woo and Sue." It was coined by Robert Thomson at News Corp a couple of years ago. The logic is brutal but effective. Step one: You sue the AI companies for copyright infringement. You make it legally terrifying for them to touch your data. You file massive lawsuits, like the New York Times did back in late 2023. [Speaker 2]: And that fear is the leverage. Because once the AI companies are scared of a trillion-dollar judgment, you move to step two: The Woo. You invite them to the table and say, "Hey, we can make all this legal trouble go away... for fifty million dollars a year." [Speaker 1]: And clearly, it worked. The checkbooks opened. OpenAI paid News Corp a reported two hundred and fifty million over five years. Google is paying Reddit sixty million a year. And now Meta has fallen in line. But here’s the thing-this strategy only works if you can actually stop the theft. If the AI companies can just steal your data for free, they’ll never pay you. So the publishers had to get technical. [Speaker 2]: This is where the arms race started. Back in the day-and by back in the day, I mean 2024-websites used a polite little text file called "robots.txt." It was essentially a sign on the door that said, "Please do not enter." [Speaker 1]: And the AI companies respected that? [Speaker 2]: Well, some did. But the data from tollBit shows that about thirteen percent of AI agents just ignored it. They walked right past the sign. So publishers realized that being polite wasn't protecting their product. They had to get aggressive. [Speaker 1]: So what does aggressive look like? Because we’re seeing reports that seventy-nine percent of top news sites are now blocking AI.…