The Great Lockout

Developer Mark Russo’s attempt to navigate the web’s new data walls results in a catastrophic lifetime ban from the digital economy.

[Speaker 1]: As of this month, January 2026, seventy-nine percent of the top news websites in the US and UK are now actively blocking AI crawlers. [Speaker 2]: Think about what that actually means. For thirty years, the internet operated on an "opt-out" basis. If you put something online, it was public unless you specifically asked robots to stay away. That era is effectively over. We’ve shifted to a "permission-based" ecosystem. [Speaker 1]: We’re tracking the emergence of a two-tiered information society. On one side, you have the tech giants who can pay to access real-time truth. On the other, you have everyone else, left with data that is increasingly stale, synthetic, or legally dangerous. [Speaker 2]: And to see where this leads, we have to look at an independent developer named Mark Russo. He tried to build an AI tool outside of the paid gates, using open data. And instead of a breakthrough, he ended up with a lifetime ban from the digital economy. [Speaker 1]: It’s Wednesday, January 28, 2026, and you’re listening to The Angle. [Speaker 2]: So to understand how we got to seventy-nine percent of the news going dark, we have to look at the massive consolidation that just happened. [Speaker 1]: Right, the deal announced earlier this month between Google and Apple. They’re merging Gemini into Siri. It’s an integration that basically cements the pipeline: Apple provides the users, Google provides the intelligence. But the fuel for that intelligence-the actual news and data-is harder to get than ever before. [Speaker 2]: Because the people who own that data-the publishers-have realized they were being robbed. If you rewind to May 2024, there was this brief moment of optimism. We called it the "Grand Bargain." News Corp signed a deal with OpenAI worth over two hundred and fifty million dollars. The theory was simple: AI is just the new search engine. It’ll drive traffic to our articles, we’ll make ad money, everyone wins. [Speaker 1]: But the math didn’t hold up. By early 2025, we had the actual numbers, and they were brutal. Traditional Google search sends about eight or nine percent of traffic through to publishers. AI search engines? They were yielding a click-through rate of 0.74 percent. Less than one percent. [Speaker 2]: That’s the collapse right there. AI wasn’t a traffic driver; it was a substitute. It reads the news so you don’t have to. Once publishers saw that 0.74 number, the "Grand Bargain" died. They realized they were training their own replacement for free. [Speaker 1]: So the walls went up. But this wasn’t just publishers updating a text file on their servers. The infrastructure of the web actually changed to help them do it. [Speaker 2]: Exactly. This is where Cloudflare comes in. In July 2025, they pivoted their entire business model. They introduced a feature that let websites block all AI bots with a single click. But more importantly, for new domains, they made "closed" the default setting. [Speaker 1]: That’s a massive shift in philosophy. It turned the internet into a toll road. And we saw the first major casualty of this mindset a month later, on August 12, 2025. That was the day Reddit blocked the Internet Archive. [Speaker 2]: That one really shook the research community. Reddit was trying to stop AI companies from scraping their conversations for training data. But to do that, they had to block everything-including the Wayback Machine. [Speaker 1]: It proved that in this new economy, history is just collateral damage. If you can’t monetize the data, you lock it down. And…