Big Tech Is Trying to Sneak AI Regulation Through Congress by Attaching It to Child Safety Laws

The AI industry's Washington lobbyists have a problem: they want a single, federal set of rules governing artificial intelligence across the entire country — what they call "preemption" — but Congress isn't interested. So they're trying a different approach: bundling AI preemption with children's online safety legislation and hoping nobody notices the fine print.

If they succeed, the consequences will touch every American who uses the internet.

## What Is Preemption and Why Does Big Tech Want It?

Right now, AI regulation in the United States is a patchwork. Different states are passing different laws — Colorado has one approach, California has another, Texas is working on its own. For Big Tech companies building AI products, this means complying with dozens of potentially conflicting state rules, which is expensive and legally risky.

Preemption would solve this by establishing one federal standard that overrides all state laws. For companies like Google, Microsoft, and Meta, this is the holy grail: one set of rules to follow, ideally written with significant industry input.

The problem? Most Americans and many lawmakers think state-level experimentation is exactly how democracy should work. Let Colorado try something, see what happens, and let other states learn from it. Preemption would kill that process dead.

## The New Strategy: Tie It to Kids' Safety

According to reports this week, the White House has told child safety groups and Big Tech companies that it would endorse a slate of children's online safety laws backed by Sen. Marsha Blackburn (R-TN), co-author of the Kids Online Safety Act (KOSA), as part of a broader preemption package.

The logic is cynical but effective: who's going to vote against protecting children online? If AI preemption is bundled with KOSA, any politician who opposes preemption can be portrayed as opposing child safety. It's a shotgun wedding between two unrelated policy priorities, and Big Tech is playing matchmaker.

But there's a fundamental mismatch. Online child safety is a real and pressing issue, but it's only one facet of AI regulation. A truly comprehensive law would need to address frontier model safety, algorithmic discrimination, environmental impact of AI data centers, labor displacement, and surveillance — issues that have nothing to do with whether a 14-year-old sees harmful content on Instagram.

## The Republican Civil War

The preemption fight has exposed a deep rift in the Republican Party. GOP leadership, aligned with Big Tech donors, wants federal preemption. Populist Republicans, energized by the success of state-level legislation, want to preserve states' rights to regulate AI. President Trump has called for an AI preemption bill, putting his party in the awkward position of needing to deliver something the base is skeptical of.

The White House is now trying to finesse a preemption approach influenced by Mike Davis, a Trump-allied lawyer who successfully killed a different AI moratorium in the Senate last year. But the internal confusion is palpable.

"No one knows really who's driving this thing," a Republican lobbyist for a midsize tech company told The Verge. "Everyone is deeply, deeply, deeply skeptical of [the bill's] movement, because everyone is on such different pages. I think the House is not going to move anything that Blackburn wants."

The White House apparently didn't inform House Republicans — who had passed their own version of KOSA — that it was going with Blackburn's legislation. Senate Democrats who'd worked with Blackburn were also left out of the loop. There's even a separate, bipartisan AI preemption bill floating around the House. The result has been a week of total confusion.

## Why This Matters Beyond Washington

If Big Tech gets preemption through Congress, here's what happens:

**State AI laws get wiped out.** Colorado's landmark AI discrimination law, California's AI transparency requirements, Illinois's AI hiring regulation — all overridden by whatever federal standard emerges. And if history is any guide, the federal standard will be weaker than the strongest state laws.

**Consumer protections get weaker, not stronger.** Preemption doesn't add protections — it removes them. The federal floor becomes the ceiling, and states can't go higher.

**The regulation window closes.** Once a federal preemption law is passed, it becomes extraordinarily difficult to amend or strengthen it. The AI industry knows this. They're not seeking a starting point for regulation — they're seeking a ceiling they can live under for the next decade.

**Kids' safety becomes a vehicle, not a priority.** Tacking AI preemption onto KOSA doesn't make the internet safer for children. It makes preemption politically harder to oppose. The child safety provisions may pass, but the AI preemption provisions will have been smuggled through on their coattails.

## What This Means For You

**Pay attention to how your representatives vote on any "comprehensive" AI bill.** The title may say "children's safety" — the fine print may say "corporate immunity from state regulation." These are not the same thing.

**State-level AI protections are your strongest current defense.** If you live in a state that has passed or is considering AI regulation, those protections are at risk of being overridden by federal preemption. Contact your state legislators and ask them to oppose preemption language.

**The "one national standard" argument sounds reasonable, but it's a trap.** When an industry asks for one set of rules, they're asking for the weakest possible set of rules applied everywhere. State-level experimentation has historically been how America figures out what works — on environmental law, consumer protection, labor standards, and now AI.

**The timeline is tight.** Midterm elections change the political calculus. If Republicans lose the House, the window for preemption closes entirely. Big Tech knows this, which is why the push is frantic now. The urgency is real — but it's urgency for their bottom line, not for your safety.

The fundamental question isn't whether AI should be regulated. It's who gets to write the rules: 50 states experimenting and competing, or one federal standard negotiated behind closed doors by the companies being regulated. The answer to that question will shape the next decade of American technology policy.