Can Congress Govern Disruptive Technology?

August 10, 2026 - 06:32

Picture this: you wake up to a frozen bank account because an AI system flagged you as a fraud risk. No human picks up the phone. Another algorithm reviews your appeal. Hours pass. The mistake gets fixed, but the trust is already broken. Who is really in charge here?

That question sits at the heart of a growing debate in Washington. Lawmakers are scrambling to understand technologies that evolve faster than legislation can be drafted. Social media platforms, facial recognition, autonomous vehicles, and generative AI all pose the same core problem: by the time Congress holds a hearing and writes a bill, the technology has already moved on.

The gap is not just about speed. It is about expertise. Most members of Congress are lawyers, former business owners, or career politicians. Very few have engineering backgrounds. They rely on staff briefings and industry testimony, which often come from the very companies they are trying to regulate. That creates a conflict of interest. A tech firm will happily explain the benefits of its product, but it will rarely volunteer the risks.

There is also the question of jurisdiction. Which committee oversees AI? Is it Commerce, Judiciary, or Energy? The answer is all of them and none of them. Responsibilities are scattered, so no single body has a full picture. This fragmentation leads to piecemeal rules that vary by state, leaving businesses confused and consumers unprotected.

Some argue that Congress should step back and let agencies handle it. The Federal Trade Commission and the Securities and Exchange Commission have already taken action on data privacy and algorithmic bias. But agencies are also underfunded and slow to adapt. They can fine a company after the damage is done, but they cannot prevent the next crisis.

The real issue is not whether Congress can keep up. It cannot, and it never will. The question is whether it can build a system that is flexible enough to respond when things go wrong. That means sunset clauses on tech regulations, mandatory audits for high-risk algorithms, and a dedicated office of technology assessment with real teeth.

Until then, we are left with a patchwork of half-measures and reactive headlines. The bank account gets unfrozen. The apology email gets sent. But the underlying problem remains: no one is truly accountable when the machines make the calls. And that is a gap no algorithm can close.