TSA's Optional Face Scans: What Voluntary Means

Stand in a TSA line at one of the 80-plus airports now running facial comparison technology, step up to the podium, and a camera captures your face. Most travelers have no idea they just made a choice. That's the problem — and it's bigger than one federal agency's rollout strategy.

TSA's own fact sheet describes the program in language so carefully neutral it practically vibrates: "A traveler may voluntarily agree to use their face to verify their identity during the screening process by presenting their physical identification or passport." Clean. Technically accurate. And almost entirely disconnected from what actually happens when you're running late, your shoes are off, and a camera is already pointed at your face.

This is the consent paradox the agency has built — and it's worth unpacking slowly, because the consequences aren't just about airport lines.

The Default Is the Message

Here's a behavioral science fact that TSA's architects almost certainly know: when the default action in a high-stress, time-pressured environment is "step forward," opt-out rates collapse. Not because people support the program. Because they don't know there's anything to support or oppose. They're just trying to catch a flight.

McKenly Redmon of Southern Methodist University's Dedman School of Law put it plainly in a recent analysis covered by The Regulatory Review: travelers are "likely unaware that they can opt out, and signage at airports frequently uses vague terms." Redmon's argument isn't that TSA is operating illegally — it's that the passengers' ability to decline "often exists only in theory."

That phrase — exists only in theory — is doing serious work. Because TSA's counterargument is technically defensible: signage is present, no traveler is denied boarding for refusing, alternatives exist. All true. None of it adds up to meaningful consent. This article is part of a series — start with Facial Recognition Checkpoint Convergence Investig.

"These biometric screenings threaten privacy, fairness, and civil liberties." — McKenly Redmon, The Regulatory Review

Meaningful consent, in any serious due process framework, requires that refusal carries no friction penalty and that the person being asked actually understands what they're agreeing to. A camera that's already capturing your image while a TSA officer waits, and a line of impatient strangers forms behind you, is not that framework.

Scale Without Guardrails

The scale of this rollout is accelerating well ahead of the policy architecture meant to govern it. TSA is already operating facial comparison — specifically, its credential authentication technology-2 (CAT-2) scanners — at more than 80 airports, with stated ambitions toward 400-plus locations nationwide. The legislative framework for informed consent at federal checkpoints has not kept pace with that expansion. Not even close.

A 2023 GAO review flagged accuracy concerns with the program and recommended stronger performance evaluation. TSA disputed those findings. That back-and-forth between the agency's assurances and independent scrutiny is now a documented credibility gap — which is exactly the kind of gap that festers into full legislative backlash when the right news cycle hits.

And here's where it gets interesting for anyone who uses facial analysis professionally, outside of airports and entirely unconnected to TSA's operation. When Congress decides to draw lines around facial technology — and it will — it draws them broadly. The technical difference between a mass-screening checkpoint camera and a forensic investigator running a precise, documented, two-image comparison is enormous. But that distinction survives only if practitioners are already on record making it, loudly, before the headlines get written for them.

The Precision Distinction That's Getting Buried

Let's be specific about what TSA is actually doing, because the public conversation has almost entirely lost this detail. CAT-2 scanners capture a real-time image and compare it against the government-issued ID the traveler is presenting. That's facial comparison — two images, both known, one purpose. It is not the same as running an unidentified face against a database of millions to generate a list of candidates. Those are different technologies, different legal exposures, and different ethical weight classes.

TSA says it deletes the captured photos except in limited cases. That assertion, accurate or not, is doing a lot of work to maintain public trust in a program that has no independent verification mechanism the public can actually see. Previously in this series: Government Grade Facial Recognition Security Risks.

The broader concern — and the one that should focus attention for anyone doing legitimate investigative work with facial comparison tools — is that public panic is flattening this distinction entirely. Every airport controversy, every congressional hearing, every civil liberties filing against biometric screening gets mentally filed under "facial recognition bad." That's the reputational blast radius. It hits programs that were built carefully and programs that weren't, because the headlines don't have room for technical footnotes.

For investigators using facial comparison methodology in real case work, this is the environment taking shape around them. The question isn't whether the TSA controversy is fair to the technology. It's whether investigators' tools are built to survive the scrutiny that's coming — precise outputs, documented methods, clear scope, explainable results. That's what separates a defensible finding from a liability when a defense attorney or oversight committee starts asking questions.

Tools like CaraComp were designed for exactly this environment — investigator-controlled, methodologically transparent, built for individual case accountability rather than population-scale screening. That distinction matters now more than it ever has.

What "Actually Optional" Would Have to Look Like

This is the question that exposes everything. If TSA wanted to run a genuinely voluntary program — not technically-defensible-voluntary, but actually-voluntary — what would that checkpoint look like?

Separate lanes, clearly marked, with equal throughput times. Written acknowledgment of the specific technology being used and what happens to the captured image. A staffed human alternative with zero added friction for travelers who decline. Signage that says "you can skip this" in plain language, not "this process is optional per federal regulation 49 CFR." Probably an independent audit of opt-out rates published annually so the public can see whether the voluntary architecture is actually working.

None of that exists at scale. And the gap between the program as described and the program as experienced is precisely where the legal and political exposure lives. Up next: Facial Tech Is Now Infrastructure Casework Still A.

"TSA introduced facial comparison technology into the screening process at select airports. The facial comparison technology represents a significant security enhancement and improves passenger convenience." — TSA.gov, Facial Comparison Technology Fact Sheet

"Significant security enhancement and improves passenger convenience." That's the official framing. It may even be accurate. But a program can improve security and convenience while simultaneously having a consent architecture that doesn't hold up — those aren't mutually exclusive. And the agency acting as if the first two qualities answer concerns about the third is exactly the kind of response that turns a manageable controversy into a legislative crisis.

TSA's program will keep expanding. The airports will multiply toward that 400-plus target. The consent debate will get louder, not quieter, and at some point a high-profile misidentification or a particularly damaging congressional hearing will snap the public conversation from slow simmer to full boil.

When that happens, the investigators and analysts who built their practice around documented, explainable, comparison-only methodology will be in a defensible position. Everyone else will be explaining why their tool is different from the thing on the front page — to an audience that stopped listening to technical distinctions about three news cycles ago.

So here's the question worth sitting with: if you walked up to a TSA checkpoint tomorrow and genuinely didn't know your face was being compared to your ID until after it happened — did you consent? And if the answer is some version of "technically, yes" — what does that tell you about how much the word "voluntary" is actually worth?