Here's an unpopular take in a town obsessed with speed: the regulatory impulse to move quickly on AI governance may be the single biggest threat to actually getting it right.
Everyone from Capitol Hill to state legislatures to international bodies seems to be operating under the same assumption: AI is moving so fast that regulators must sprint to keep up. The logic is intuitive. The technology evolves at breakneck pace. Waiting means falling behind. Falling behind means ceding control to industry. So regulate now, fix it later.
Except that's backwards. Fast regulation on emerging technology tends to calcify mistakes into law. It privileges whoever got to policymakers first. And it often creates the exact opposite of what regulators intended.
Look at what happens when governments rush. They tend to regulate what they can see and understand in the moment, which usually means the previous generation's problems. They miss second-order effects. They write rules so specific to current technical implementations that they become obsolete within months, forcing expensive compliance with rules that no longer make sense. Worst of all, they often entrench the largest players by building compliance infrastructure that only big companies can afford to navigate.
The current moment feels urgent because it is. Real harms exist. The invisible instruction injection in court filings that was recently revealed, the systemic risk warnings from Treasury analysts, the growing evidence of manipulation through automated systems—these aren't hypotheticals. They're happening now. That urgency is legitimate.
But urgency and haste are not the same thing.
The more productive question isn't "how do we regulate this faster?" It's "what do we actually need to understand before we regulate this at all?" That's a different kind of work. It's slower. It's messier. It requires sustained technical expertise, not soundbite-friendly legislation. It means admitting uncertainty and building mechanisms for adaptive oversight rather than writing permanent rules.
Consider what we're seeing with various regulatory proposals: many focus on transparency requirements, disclosure rules, and algorithmic auditing. These are sensible starting points. But they're also the easy targets—the regulations that sound good in a hearing and seem technically feasible. They're not necessarily the regulations that would actually address the deepest concerns. And because they're easier to write, they tend to get passed first, creating the false impression that the regulatory job is done.
Meanwhile, the genuinely hard questions go unresolved. How do you regulate systems whose decision-making is genuinely inscrutable even to their creators? How do you balance innovation with protection when you don't yet know what you're protecting against? How do you write rules that apply internationally when different jurisdictions have fundamentally different values? These questions don't have quick answers.
The risk of rushing is particularly acute around questions of access and concentration. When regulators move fast, they often accidentally entrench market leaders. Licensing requirements sound good in theory but favor companies with legal departments. Safety certifications sound reasonable until you realize they cost millions to implement. Interoperability mandates sound pro-competition until you see them implemented in ways that only benefit incumbents with existing infrastructure.
None of this is an argument for inaction. It's an argument for distinguishing between genuine urgency and the pressure to appear busy. Emergency measures should be genuinely temporary and narrowly targeted. Broader regulatory frameworks should be allowed to take time. We should fund research to understand these systems better. We should run pilot programs. We should let different jurisdictions experiment before settling on global approaches.
Speed has its place. Emergency responses to imminent specific harms deserve swift action. But the instinct to regulate quickly across the board is a form of false urgency—the kind that feels responsible but often produces worse outcomes than patience.
The harder thing, politically and institutionally, is to move deliberately. But it's usually the right call.