AI Airport Body Scanners: How They Work Now
Yes, the United States is actively using artificial intelligence in airport body scanners, primarily to speed up screening and reduce human error. The technology, deployed by the Transportation Security Administration (TSA) at many checkpoints, uses machine learning to automatically detect potential threats on a passenger’s body, replacing the need for an officer to interpret the generic outline image. This article explains what the AI actually does, how it changed the screening process, and what it means for your next flight.
What the AI Actually Does
Since around 2018, the TSA has been deploying body scanners that use automated target recognition (ATR) software. The older generation of scanners showed a generic, chalk-like figure on a screen, with yellow boxes marking areas that needed a manual pat-down. A human officer had to interpret that image and decide whether the highlighted area was a threat or a harmless fold in clothing. The new AI-based systems do that interpretation automatically. The scanner’s software compares the sensor data against a large library of images of both threatening and benign items, then decides in a fraction of a second whether to clear the passenger or flag a specific area for a secondary check. The officer never sees a body image at all; they only see a message like “no threats detected� or a simple outline with a box indicating where to pat down.
Background: From Naked Scanners to AI
The current AI scanners are the third generation of this technology. The first generation, introduced in the late 2000s, used millimeter-wave or backscatter X-ray technology and displayed a detailed, almost nude image of the passenger. That raised serious privacy concerns, so the TSA switched to the “generic avatar� version, where the officer saw a standardized figure with threat boxes. But even that required an officer to judge whether the box was a real threat. The AI version removes that human judgment step entirely. It was rolled out gradually, starting with a few airports in 2018, and by 2021 the TSA had deployed the new systems to most major checkpoints. The agency has stated that the AI reduces false alarm rates and speeds up the line, though exact performance figures are not publicly released.
How the AI Scanner Works
The scanners use millimeter-wave technology, which emits low-power radio waves and reads the reflections. The data is then processed by a neural network, a type of AI trained on millions of scans. The training set includes images of weapons, explosives, and everyday items like wallets, belts, and medical devices. The network learns to distinguish between a harmless object and a threat based on shape, density, and position. When you step into the scanner, it takes a 360-degree scan in about two seconds. The AI then classifies any anomalies. If it finds something that matches a threat pattern, it marks the location on a generic figure. If not, you’re cleared. The system is designed to ignore body features like breasts, genitals, and prosthetics, which is why it can operate without showing an image to the officer.
Privacy and Safety: Common Misconceptions
Many travelers still worry that the scanner “sees� them naked. That is not true for the current generation. The AI processes the data and discards the raw image immediately; it is never displayed or stored. The system is also safe: millimeter-wave scanners emit non-ionizing radiation, similar to a cell phone, and the FDA has approved them for use on all people, including pregnant women and children. The TSA says the scanners meet all national safety standards, and the radiation exposure per scan is far below the limit for a single X-ray. Another misconception is that the AI can detect items hidden inside body cavities. It cannot; it only detects items on the surface of the body or in clothing. That’s why pat-downs are still used when a passenger is flagged.
How to Choose: What to Know as a Passenger
You cannot choose which type of scanner you go through; the airport decides based on what they have installed. However, you can choose to opt out of the body scanner entirely and request a pat-down instead. This is your legal right under TSA policy. If you have a medical implant, a prosthetic, or a condition that might be affected by the scanner, you can also request a pat-down. Some passengers with pacemakers or insulin pumps worry about interference, but the TSA states the scanners are safe for all medical devices. If you are concerned, just tell the officer you’d like a pat-down. You will be screened in a private room if you request it, and the officer will use a metal detector wand and a manual pat-down.
Common Problems and What to Do
The most common issue is a false alarm, where the scanner flags a fold in clothing, a button, or a piece of tissue. If that happens, you’ll be asked to step aside for a pat-down. To reduce the chance, empty your pockets completely, remove belts, watches, and jewelry, and make sure your clothing has no large metal zippers or snaps. Another issue is that the scanner sometimes flags medical items like a colostomy bag or a catheter. In that case, you can request a private screening and explain the situation. The officer is trained to handle these cases with respect. If you feel the process was invasive or handled improperly, you can file a complaint with the TSA through their website or by phone.
Comparison: AI vs. Older Scanners vs. Metal Detectors
The AI scanners are a significant upgrade over the older ATR scanners and the traditional metal detectors. Metal detectors only find metal, so they miss ceramic knives, plastic explosives, and liquids. The older ATR scanners found those items but required an officer to interpret the image, which could lead to errors. The AI scanner finds both metal and non-metal threats and makes the decision automatically, removing human error. The trade-off is that the AI is only as good as its training data; it might miss a novel threat that looks nothing like anything in its database. That’s why the TSA still uses random pat-downs and behavioral detection. In practice, the AI scanner is faster, more private, and more accurate than its predecessors.
Costs and Implementation
The TSA does not publicly disclose the exact cost per unit, but the agency has spent hundreds of millions on the deployment. A single millimeter-wave scanner costs roughly $100,000 to $200,000, and the AI software is an additional licensing fee. The TSA has installed these scanners at over 200 airports nationwide, with plans to replace all older models. The cost is covered by the federal budget, not by passenger fees. The main operational cost is training officers to use the new interface and maintaining the software. The AI models are updated periodically to improve detection and reduce false alarms.
Safety and Accuracy: What the Data Shows
The TSA has not released detailed performance statistics, but independent studies and government reports indicate that the AI scanners have a lower false alarm rate than the older ATR systems. A 2020 report from the Government Accountability Office (GAO) noted that the new scanners reduced the time needed for screening and improved privacy, but also found that the TSA had not fully validated the AI’s performance in all scenarios. The scanners are tested regularly by TSA’s internal testing lab, which uses mock threats. The system is designed to detect a wide range of items, including guns, knives, explosives, and even 3D-printed firearms. However, no technology is perfect, and the TSA still relies on multiple layers of security.
Frequently Asked Questions
Can the AI scanner see my body? No. The current generation does not display or store any image of your body. The AI processes the sensor data and only outputs a generic figure with threat boxes if something is found.
Is the scanner safe for pregnant women? Yes. The millimeter-wave technology uses non-ionizing radiation, similar to a cell phone, and the FDA has approved it for all people, including pregnant women.
Can I opt out of the body scanner? Yes. You can always request a pat-down instead of going through the scanner. Just tell the officer at the checkpoint.
Does the AI scanner detect drugs? It is not designed to detect drugs. It is focused on weapons and explosives. If you have a legal medical substance in a pill bottle, it will not be flagged unless it looks like a threat.
What happens if the scanner flags me? You will be asked to step aside for a pat-down. The officer will check the flagged area. You can request a private screening at any time.
Related
Method: the standard formula for this calculation, rounded as each result states. See how the calculators work or report an error.