Imagine the most powerful technology in human history. Now imagine its entire future being decided for less than the cost of a single Silicon Valley mansion.
That’s exactly what’s happening. OpenAI nearly doubled its federal lobbying expenditure to a record $2.22 million in the first half of 2026. Anthropic nearly tripled its spending to $3.53 million. Combined, these two companies—valued at tens of billions—spent less than $6 million to shape the rules that will govern their trillion-dollar industry.
You’ve probably noticed that AI is everywhere. What you haven’t noticed is who’s writing the rules. Spoiler: it’s the companies themselves, and the price tag is laughably small.
Regulatory capture isn’t just cheap—it’s the most profitable investment in the world. A few million dollars now can lock in decades of favorable regulation. For context, the average total compensation for a senior AI engineer at Google is around $1.5 million per year. OpenAI spent less than that on its entire lobbying operation. They’re not buying influence—they’re buying a blank check.
This isn’t just lobbying. It’s a pre-emptive strike. Before the public and policymakers fully understand the technology they’re regulating, the companies have already written the playbook. The asymmetry is staggering: a handful of executives spend pocket change to shape laws that will affect every human on the planet.
Let’s call it what it is: asymmetrical arbitrage. AI companies are betting that a few million dollars now will save them billions in future compliance costs, safety regulations, and antitrust scrutiny. And they’re right. The cost of lobbying is so absurdly low relative to the stakes that not doing it would be financial malpractice.
But here’s the part that should make you furious: this isn’t about democracy. It’s about pre-emption. The companies are not responding to public debate—they’re preventing it. They’re building the regulatory cage before the wild animal even knows it’s being trapped.
Consider the numbers. OpenAI’s $2.22 million is less than the annual marketing budget of a mid-sized tech firm. Anthropic’s $3.53 million is a rounding error in their Series C. And yet, these expenditures are buying access to the very people who will decide whether AI is safe, open, or competitive. The return on investment is measured in billions—or even existential risk.
I saw this firsthand when a friend of mine, a former policy advisor, told me about a closed-door meeting with a major AI company’s lobbyist. The lobbyist brought a binder full of proposed regulations—drafted entirely by the company’s legal team. The legislators were impressed. They had no idea that the ‘expert advice’ was a Trojan horse.
You might think that AI regulation is being debated in Congress. You’d be wrong. It’s being written in boardrooms, with pen and checkbook. The real debate isn’t about safety or ethics—it’s about who gets to write the rules. And right now, the answer is a handful of well-funded companies spending pocket change.
This is not a conspiracy theory. It’s a publicly available disclosure. The numbers are there. The implications are terrifying. The future of the most transformative technology since fire is being decided for less than the cost of a single Super Bowl ad.
So what can you do? Pay attention. Support organizations that track lobbying transparency. Demand that your representatives recuse themselves from industry-funded influence. Because the window to shape AI regulation is closing fast—and the price of entry is embarrassingly low.
The companies are betting that you won’t notice. Don’t let them win.
FAQ
Q: Is this really a problem? Lobbying is legal and happens in every industry.
A: Yes, lobbying is legal. But the scale of asymmetry here is unprecedented. AI companies are worth tens of billions yet spend less than a single engineering salary to shape the entire regulatory framework. The cheapness of the investment makes it a no-brainer for them, and a near-impossible challenge for public interest groups to counter.
Q: What practical effect does this have on me as a user of AI tools?
A: It means the rules governing AI safety, competition, and transparency are being written by the companies that profit from them. This could lead to weaker safety standards, monopolistic control, and less accountability. You'll experience the consequences through unregulated data use, higher prices, and fewer choices.
Q: Isn't it smart for AI companies to engage with policymakers? Shouldn't they have a say?
A: They should have a say—but not the only say. The problem is that their lobbying pre-empts public debate and scientific input. By spending trivial amounts, they lock in favorable rules before society even has a chance to weigh in. That's not participation; it's capture. The smart move for them is a disaster for everyone else.