Let me start by asking you this: What if the people who hold the power to shape public discourse were suddenly forced to defend their work in a courtroom where the judge was an algorithm? That’s the surreal experiment Peter Thiel’s latest venture, now called The Primary, has been testing. But here’s the kicker—this isn’t about holding journalists accountable. It’s about creating a new currency for credibility, one that might redefine the very soul of investigative reporting. And honestly, I think this is both brilliant and terrifying in equal measure.
The story begins with Aron D’Souza, the man who once weaponized legal loopholes to dismantle Gawker. His new project, Objection, initially aimed to let AI-powered tribunals arbitrate disputes between journalists and their subjects. The idea was simple: If a billionaire felt wronged by a reporter’s exposé, they could pay to have the story reviewed by an AI system. But the experiment hit a wall almost immediately. As D’Souza himself admitted, the AI couldn’t handle the nuance of human judgment. It wasn’t about justice—it was about slop. And that’s when the pivot happened.
Now The Primary exists, not as a courtroom, but as a digital scoreboard. It ranks journalists and outlets based on metrics like source attribution, tone, and whether they give subjects a chance to respond. The twist? This isn’t just another ratings system. It’s a radical reimagining of how journalism is valued. Personally, I think this could either be the salvation of investigative reporting or its slow death knell. Why? Because if credibility becomes a number, what happens to the messy, unquantifiable work of digging up truths that no one wants to hear? That’s the question no one seems to be asking loud enough.
Let’s talk about the implications. D’Souza’s team claims their AI model can finally give reporters a ‘portable, public record’ of their work. Imagine a journalist walking into a newsroom with a score that proves they’ve done the heavy lifting—sourced stories, avoided bias, and given everyone a fair shot. Sounds ideal, right? But here’s the rub: What if this system ends up rewarding the same old gatekeepers? The Washington Post and CBS News currently lead the rankings. That’s not a coincidence. If the algorithm favors established outlets, it could entrench the existing power structures rather than disrupt them. And that’s a problem because the real issue with journalism isn’t the quality of reporting—it’s who gets to do it in the first place.
What makes this particularly fascinating is the economic angle. D’Souza argues that the current system rewards attention, not accuracy. Awards are given to a tiny fraction of reporters, while the rest are left in the dark. But what if the true value of journalism is measured by something other than clicks or accolades? The Primary’s model suggests that credibility itself could become a tradable asset. That’s a dangerous idea, though. If a reporter’s worth is reduced to a score, what happens to the stories that don’t fit neatly into algorithmic criteria? The ones that require months of chasing leads, interviewing sources no one else wants to talk to, and risking everything to expose corruption? Those stories might not get ranked at all.
And then there’s the question of who controls the algorithm. The white paper from The Primary outlines metrics like ‘source attribution’ and ‘coverage tone,’ but who decides what those terms mean? If the system is trained on data from existing power structures, it could reinforce biases rather than challenge them. I’m not saying this is inherently bad, but I am saying it’s a massive gamble. The founders claim they’ve learned from their tribunal days—that enforcement only works on the margins. But what if the scoreboard becomes the new enforcement mechanism? A journalist with a low score could find themselves blacklisted, their work ignored, their career derailed. That’s not justice—it’s a new kind of censorship, one that’s algorithmically enforced.
Let’s not forget the irony. D’Souza’s own experience with AI systems digesting his interview with TechCrunch highlights the core contradiction: The reporter does the work, but the machine takes the value. That’s the economics of journalism in a single afternoon. And yet, The Primary’s solution is to build a system that measures the value of that work. It’s a noble goal, but it’s also a deeply flawed one. What if the act of measuring something changes what it is? If journalists start tailoring their work to game the algorithm, we might end up with a generation of reporters who prioritize scoring over substance. That’s not progress—it’s a dystopia waiting to happen.
In the end, The Primary is less about fixing journalism and more about rebranding it. It’s a bold experiment, one that could either democratize credibility or deepen the divide between the powerful and the powerless. What I find most interesting is the way it reflects a broader trend: We’re increasingly outsourcing our trust to machines. But trust isn’t something that can be quantified. It’s earned through years of hard work, integrity, and the willingness to say, ‘This story matters, even if no one wants to hear it.’ If The Primary’s model succeeds, it might change how we value that kind of courage. But I suspect it’ll also change how we define it. And that’s a conversation worth having.
