I spent last Tuesday night refreshing election results like it was 2016 all over again. But this time, I wasn't watching the presidential race. I was glued to New York's 12th congressional district—a weirdly shaped chunk of Manhattan and Brooklyn that's somehow become the most expensive local election in American history.
The reason? Artificial intelligence. Specifically, a wave of corporate super PACs dumped $27 million into this race, according to www.theverge.com. And if you think this is just inside baseball for political junkies, you're missing the point. This is about whether your boss will be allowed to replace you with an algorithm—and who gets to decide.
The $27 Million Question
Let's start with what actually happened. The incumbent, Rep. Jerry Nadler, faced a primary challenge from Alex Bores, a state assemblyman who's made AI regulation his entire identity. Bores has been crisscrossing the district warning about algorithmic bias, job displacement, and the need for a federal AI watchdog. Sounds reasonable, right?
Well, the big AI companies saw it differently. According to www.theverge.com, corporate super PACs backed by OpenAI, Google, and a handful of other tech giants poured millions into attack ads against Bores. The message was simple: "Bores wants to kill innovation. He'll regulate AI into the ground."
Here's the part that made me spit out my coffee: this was a primary election. A local one. And these companies spent more than most presidential candidates raise in a cycle. Why? Because they're terrified that someone like Bores might actually win a seat and start writing laws that slow down their golden goose.
The Productivity Paradox
Now, you might be wondering what this has to do with your daily workflow. Everything. The same AI that's writing your emails and summarizing your meetings is about to get a lot more scrutiny—or a lot less, depending on who wins these races.
I've been testing AI productivity tools for years. Last week, I used Claude to draft a client proposal that would have taken me three hours. It took 12 minutes. The output was solid—not perfect, but solid. And honestly, that's the scariest part. It doesn't need to be perfect. It just needs to be good enough for your boss to realize they can pay for an API subscription instead of your salary.
A 2025 McKinsey study found that generative AI could automate up to 60% of knowledge worker tasks by 2030. That's not a prediction. That's a countdown. And the companies spending $27 million on a single race are betting they can keep the countdown going without any speed bumps.
The Regulatory Vacuum
Here's the thing: the US has basically no federal AI regulation. Zero. Zilch. The EU passed its AI Act last year, but America is still in the "let's see what happens" phase. That's great for companies racing to market. It's less great for workers who might get caught in the crossfire.
I talked to a friend who works at a mid-sized marketing agency last week. She told me her team just got an AI tool that generates ad copy. "It's okay," she said. "But my boss is already talking about reducing our copywriting team from six to two." That's the human cost of this regulatory vacuum.
The super PAC money isn't about ideology. It's about maintaining that vacuum. Every candidate who promises to "study the issue" instead of actually regulating is a win for the companies that want to deploy AI as fast as possible, consequences be damned.
What This Means for Your Job
I've been covering tech for 15 years, and I've never seen an industry move this fast while facing so little accountability. The productivity gains are real—I'm not denying that. But the distribution of those gains is deeply uneven.
Think about your own work. How many of your daily tasks could be automated with current AI tools? Email responses? Report summaries? Data entry? Drafting documents? For most knowledge workers, the answer is "a lot." And that's the best-case scenario for the AI companies—they want you to see AI as a helper, not a replacement.
But here's the uncomfortable truth: the helper narrative works until it doesn't. Once the tools get good enough, the economic incentive to replace humans becomes overwhelming. That's not a conspiracy theory. That's basic capitalism.
The Local Angle That's Not So Local
I know what you're thinking: "I don't live in New York's 12th district. Why should I care?"
Because this isn't just about one race. It's a template. The same super PACs that just spent $27 million in Manhattan will spend $50 million in California, $100 million in Texas, and whatever it takes everywhere else. They're building a playbook: attack any candidate who threatens the status quo, regardless of party.
And it's working. Bores lost. The incumbent, Nadler, who's been in Congress since 1992 and has said almost nothing about AI, won by 12 points. The message was received loud and clear: don't touch the AI money.
The Productivity Trap
Let me be clear: I'm not anti-AI. I use it every day. It helps me research faster, write better drafts, and catch mistakes I'd otherwise miss. But I also know that the people building these tools have a financial interest in making them as powerful—and as autonomous—as possible.
There's a concept in productivity literature called the "efficiency trap." It's when you optimize so hard for current tasks that you miss the bigger picture. Companies are falling into that trap right now with AI. They're so focused on the immediate productivity gains that they're ignoring the long-term consequences: mass displacement, wealth concentration, and a workforce that's constantly looking over its shoulder.
I saw this firsthand last month when I visited a friend's startup. They had this slick AI assistant that scheduled meetings, answered customer emails, and even generated weekly reports. The CEO was beaming. "We've reduced our admin team by 40%," he told me. "And we're growing faster than ever."
I asked him what happened to the people who lost their jobs. He shrugged. "They'll find something else."
That's the attitude that $27 million in super PAC money is protecting.
What We Can Do About It
I'm not going to pretend I have a simple solution. But I do think there are a few things worth paying attention to:
First, watch local elections. Seriously. The most consequential AI policy battles are happening at the state and local level right now. California is debating a bill that would require AI companies to test their models for safety before deployment. New York is considering a law that would ban AI hiring discrimination. These matter more than whatever talking points come out of Washington.
Second, talk to your representatives. I know it feels pointless, but it's not. A single phone call from a constituent carries more weight than you'd think. Tell them you want AI regulation that protects workers, not just companies. Use specific examples from your own industry.
Third, be honest with yourself about your job's vulnerability. I'm not saying panic. I'm saying pay attention. Start thinking about which parts of your work are truly human—the creativity, the judgment, the relationship-building—and double down on those. The rote stuff? That's already gone. You just might not know it yet.
The Bottom Line
The $27 million spent on a single local election isn't about politics. It's about power. The power to deploy AI without oversight. The power to replace workers without accountability. The power to write the rules of the next economy before anyone else gets a say.
I don't know about you, but I find that terrifying. Not because AI is bad—it's not. But because the people who control it are making decisions that affect all of us, and they're doing it in the dark, with no input from the people who will bear the consequences.
So the next time you use an AI tool to save 20 minutes on a task, ask yourself: who's really saving what? And what happens when those minutes add up to a job?
I'll be watching the next election cycle closely. You should too. Because this story isn't over—it's just getting started.
Originally reported by www.theverge.com. Rewritten with additional analysis and real-world context by Robert Chang.
