You’ve felt it. That weird tension in the Slack channel when someone mentions AI. The way the room splits — not along technical lines, but along tribal ones. The evangelists on one side, the skeptics on the other, and nobody actually listening to anybody.
Something broke in the tech community over the last two years, and it wasn’t the models. It was us.
It started with a molehill. A Linux kernel maintainer said something about AI-generated code submissions — basically, don’t dump unreviewed AI slop into the mailing list. Reasonable, right? This is Linux. Code review is sacred. Merit isn’t claimed; it’s earned through brutal, transparent scrutiny.
But within hours, that molehill became a mountain. The AI crowd screamed “Luddite.” The open-source old guard screamed “slop.” And the actual technical question — whether AI-assisted patches meet the project’s quality bar — got buried under a pile of identity politics.
The fight was never about code quality. It was about who gets to decide what counts as real.
This is what the AI gold rush is actually doing to us. Not making us smarter. Not augmenting our intelligence. It’s weaponizing our deepest insecurities — the fear of being left behind, the hunger to be on the winning team — and turning collaborative communities into zero-sum battlegrounds.
Think about it. Linux is arguably the most successful collaborative project in human history. Thousands of contributors across continents, languages, and time zones, held together by a shared commitment to merit. No marketing budget. No hype cycle. Just code that works and a culture that rewards competence.
And now that culture is being stress-tested by an industry that has more money than patience and more narrative than substance.
The AI companies need you to believe two things simultaneously: that this technology is so powerful it will reshape civilization, and that anyone who questions it is on the wrong side of history. That’s not a technical argument. That’s a sales pitch wrapped in inevitability.
When a technology can’t be questioned without the questioner being branded obsolete, that’s not progress — that’s orthodoxy.
I’ve watched this play out in real time. A senior engineer raises a legitimate concern about hallucinated code in production. Instead of engaging with the concern, someone replies with “sounds like someone’s afraid of being replaced.” The technical merit of the objection? Irrelevant. The social dynamics? Everything.
This is the dark game being played on a global scale. The AI gold rush doesn’t just exploit compute — it exploits human psychology. It manufactures enemies. The open-source maintainer who wants code review isn’t a gatekeeper; they’re doing their job. But in the gold rush narrative, they become an obstacle to progress, a dinosaur blocking the meteor.
The contradiction is staggering. AI was sold to us as a tool for augmentation — something that would amplify human intelligence, not fracture human cooperation. The pitch was “better together.” The reality is “pick a side or get steamrolled.”
And the money makes it worse. Billions of dollars are flowing into an industry that desperately needs to maintain the perception of unstoppable momentum. Doubt is the enemy of valuation. So every critique, no matter how technical, gets reframed as fear. Every question becomes heresy. Every skeptic becomes a cautionary tale the evangelists can point to and say, “Don’t be that person.”
The cruelest trick the AI industry pulled was convincing smart people that skepticism is a personality flaw.
Here’s what I know: the communities that built the infrastructure the entire AI industry runs on — Linux, Python, PyTorch, Kubernetes — were built on a culture of radical meritocracy. You bring your best work. It gets reviewed mercilessly. If it’s good, it ships. If it’s not, you try again. No hype required.
That culture is now under siege, not by bad technology, but by bad incentives. When the incentive shifts from “does this work?” to “are you on the right team?”, the foundation rots. And we all stand on that foundation, whether we admit it or not.
Linus was right. Not because AI-generated code is inherently wrong — it isn’t. But because the principle matters: code earns its place through merit, not through the enthusiasm of whoever submitted it. That principle is what made open source the most powerful collaboration model the world has ever seen.
AI is wrong not because the technology is bad, but because the gold rush around it is corroding the very cooperation that makes progress possible. It’s turning engineers into evangelists, maintainers into villains, and technical debates into culture wars.
The real threat isn’t that AI will replace us. It’s that the AI gold rush will make us forget how to work together — and that’s the one skill no model can generate.
If you’re in tech, you have a choice. You can play the tribal game, pick a side, and let the hype machine set your agenda. Or you can insist on the thing that actually matters: does the work hold up? Does the code work? Can you defend it when someone smarter than you tears it apart?
The gold rush will end. They always do. What remains is the culture. Guard it.
FAQ
Q: Isn't this just anti-AI fearmongering?
A: No. The technology itself is neutral. The problem is the gold rush economy around it — the billions in funding that require narrative control and punish skepticism. You can think AI is useful and still think the hype machine is toxic.
Q: So what should developers actually do?
A: Insist on merit. Review AI-generated code with the same rigor as any other code. Don't let hype override engineering judgment. If your team treats skepticism as a personality flaw, that's a red flag, not a feature.
Q: Isn't Linus just being a gatekeeper here?
A: Gatekeeping is only bad when it's arbitrary. Linux's gatekeeping is based on code quality and maintainability — the exact standards that made it the backbone of modern computing. Calling that 'gatekeeping' is like calling a building inspector a gatekeeper.