Facial Recognition Isn't Getting Banned. Mass Surveillance Is. Here's the Difference.

Three different governments, three different approaches to the same technology — and they're all moving at the same time. Illinois is pushing a bill that would block police from using facial recognition entirely. Meanwhile, London's Metropolitan Police say their live facial recognition cameras led to over nine hundred sixty arrests in a single year — for crimes including rape, domestic abuse, and knife attacks.

If you've ever walked past a security camera, this

If you've ever walked past a security camera, this story is about you. It's about what happens when a camera doesn't just record — it identifies. And the rules governing that capability are being written right now, in real time, on three continents at once. Illinois House Bill 5521, called the Biometric Surveillance Act, wouldn't just limit how police use facial recognition. It would bar them from obtaining, retaining, possessing, accessing, requesting, or using a biometric identification system. At the same time, the U.K. government is building a legal framework to expand live facial recognition across its cities. And China just launched its most detailed enforcement roadmap ever under its Personal Information Protection Law — with new rules that took effect 06-01-2025. So which side is getting it right — the ones banning the technology or the ones scaling it up?

Start with what's actually happening on the ground in Illinois. H.B. 5521 doesn't just say police can't run live cameras on street corners. It goes further. The bill would also prohibit law enforcement from signing agreements with outside vendors or other agencies that might preserve their access to biometric tools through a back door. That's a deliberate move to close the most common workaround — departments contracting with a third party to run searches they can't legally run themselves. A retired police chief from Riverside pushed back hard, calling facial recognition one of the most important investigative tools to come along in policing in fifty years. And opponents of the bill warn that eliminating it outright would be, in their words, reckless and dangerous. Over the past five years, at least eighteen states have considered legislation regulating law enforcement's use of this technology. But most of those states landed somewhere different than Illinois. Many of them said facial recognition can't serve as the sole basis for a law enforcement action — but they didn't ban it altogether.

That distinction matters more than any headline about bans or deployments. There are really two different tools being debated under the same name. One is live facial recognition — cameras scanning crowds in real time, trying to match faces against a watchlist. That's what London's Metropolitan Police are doing. The other is facial comparison — an investigator takes a specific photo from a specific case and checks it against a database, side by side, looking for a lead. Not a conviction. A lead that still needs corroborating evidence. Regulators around the world are starting to treat these as fundamentally different activities. And the line between them is becoming the legal boundary that determines what's allowed and what isn't.

China's new rules make this split especially clear

China's new rules make this split especially clear. The enforcement campaigns launching in 2026 target companies that collect more biometric data than necessary, fail to disclose when they share it with third parties, or use facial recognition as the only way to access a service. Notice what's being banned — not facial comparison itself, but using it as a forced default. The central rule is necessity. Facial recognition can only be used when truly required, never simply for convenience. For businesses in China, that means you can't make someone scan their face just to enter a gym or pick up a package if there's another reasonable option. For anyone who's ever been asked to verify their identity with a selfie — and wondered whether that was really necessary — China just answered that question with a regulation.

The U.K. sits on the opposite end. The government there described facial recognition as, and I'm quoting, a valuable tool to modern policing. Those nine hundred sixty-two arrests by the Metropolitan Police between September 2024 and September 2025 are being used as political validation for expanding live deployment. But even the U.K. government acknowledged that scaling this up requires — their words — a more specific legal framework. They're not handing police an unlimited license. They're trying to build guardrails around a tool they've already decided to keep.

And there's a wrinkle critics of the Illinois ban keep raising. If a state bans police from using facial recognition entirely, the technology doesn't disappear. It shifts to federal agencies that aren't bound by state law. Or it moves underground, where there's no oversight at all. The argument is that guardrails create accountable systems, while absolute bans just push usage into spaces with even less transparency. For a parent whose child goes missing, the question isn't abstract. It's whether the detective working the case has access to a tool that might help — and whether anyone's watching how that tool gets used.