Does Hertz at LAX use AI car damage scanners?
Probably yes, but Hertz has never published an official location list — which is the actual problem. Hertz confirmed 10 US airports as of August 2025 and stated intent to reach roughly 100 airport locations. Traveler-maintained lists include LAX, but that's crowdsourced, not confirmed. Assume the scanner is there and photograph the car yourself.
Why — the first-principles explanation
Rental damage disputes have always been a measurement problem disguised as an honesty problem. A tired agent walks around the car in 40 seconds under sodium lights and marks an X on a diagram. That process misses small damage constantly — which cuts both ways: the company eats real damage it can't prove, and you get blamed for scratches from three renters ago. Nobody was happy, but everyone understood the rules.
UVeye's scanner changes one variable, and it changes everything downstream. It photographs the body, glass, tires, and undercarriage in high resolution in seconds, then compares your return against your pickup. The detection floor drops from "what a person notices while walking" to "what a camera array resolves." And here's the key insight: this doesn't just find damage more accurately — it finds a category of damage that previously didn't functionally exist. Curb rash on one wheel was always physically there. It just never used to be an event. Now it's a bill.
So the technology is doing exactly what it says, and people are still furious. The reason is that the industry's fee structure was calibrated for a system that missed things. The Hertz cases that went viral weren't about false detections; a $250 repair plus $125 processing plus $65 admin, arriving automatically for a scuff you'd never have been charged for, is a policy outcome, not a computer vision outcome. When you make a detector more sensitive without recalibrating the threshold for action, you don't get fairness — you get more charges. That's the whole story.
Which is why the deployment secrecy matters more than the scanner. Hertz publicizes the technology but not which lots have it. That asymmetry means you can't know whether to document your car at pickup — and documentation is the only defense that works. The rational move is to assume every major airport has it and behave accordingly, because the cost of being wrong is a $440 bill and the cost of being careful is 90 seconds.
An example that makes it click
Imagine your landlord swapped the move-out walkthrough for a forensic lab. Under a blacklight, every apartment you've ever left has stains. The lab isn't lying. Those stains were always there.
But your security deposit was priced for a landlord glancing at the carpet. Nobody renegotiated the deposit when the blacklight showed up — they just started keeping it. That's the rental car scanner. The camera is honest. The fee schedule never got updated to match what honesty now reveals.
How to do it
- Assume the scanner is present at any major US airport Hertz location, including LAX. Hertz has not published an official list, so absence of confirmation is not confirmation of absence.
- Before you leave the lot, take a slow walkaround video with your phone. Narrate the date and location. Get all four wheels close up — curb rash is the single most common automated charge.
- Photograph the undercarriage area, windshield, and roof if you can. The scanner covers glass and undercarriage; a human agent never did, so these are the new surfaces of dispute.
- Do the same walkaround at return, before you hand over the keys.
- If you're billed, dispute it in writing immediately and request the actual before-and-after scan images. The comparison is the entire basis of the charge; make them show it.
- If Hertz doesn't resolve it, dispute the charge with your card issuer and file a complaint with your state attorney general's consumer protection office. Congress has also opened an inquiry into AI-based rental damage assessment.
Key facts
- As of August 26, 2025, Hertz confirmed the UVeye system was in use at 10 US airports, with stated plans to roll it out nationwide; reporting indicates a target of roughly 100 airport locations.
- Hertz has not published an official list of scanner-equipped locations; traveler-maintained lists circulating on forums include LAX, but these are crowdsourced and unverified.
- Of over 675,000 Hertz rentals scanned as of August 2025, more than 97% showed no billable damage — meaning roughly 3% or fewer resulted in charges.
- The UVeye system captures 360-degree high-resolution images of body, glass, tires, and undercarriage at both pickup and return, then uses AI to compare them.
- A widely reported Atlanta case produced a $440 automatic bill for curb rash on one wheel: $250 repair, $125 processing, $65 administrative.
- The US House Committee on Oversight and Government Reform opened an investigation into the use of artificial intelligence for car rental damage assessments.
- Avis Budget Group has tested AI inspection at select facilities including LAX but states its inspections remain human-led — the AI-versus-human review distinction, not the camera, is the contested issue.
▶ The 60-second explainer (script)
Does Hertz at LAX have the AI scanners? Probably — and the fact that I have to say 'probably' is the actual problem. Hertz confirmed ten US airports as of August 2025 and said it's going nationwide, targeting around a hundred locations. Traveler-maintained lists include LAX. But Hertz has never published an official list. Here's what's really going on. Rental damage disputes were always a measurement problem. A tired agent walks around your car in forty seconds under orange lights and marks an X on a diagram. That misses stuff constantly — which cuts both ways. You got blamed for scratches from three renters ago; they ate real damage they couldn't prove. UVeye's scanner photographs body, glass, tires, and undercarriage in high resolution and compares your return to your pickup. So the detection floor drops from 'what a person notices while walking' to 'what a camera array resolves.' And here's the key thing: it doesn't just find damage more accurately. It finds a category of damage that previously didn't functionally exist. Curb rash on a wheel was always physically there. It just was never an event. Now it's a bill. Hertz says over ninety-seven percent of scans show no billable damage. Fine. But the viral cases weren't false detections — a two-fifty repair plus one-twenty-five processing plus sixty-five admin, arriving automatically for a scuff, is a policy decision, not a computer vision result. You made the detector more sensitive without recalibrating the threshold for action. That doesn't produce fairness. It produces more charges. So: assume it's there. Film your walkaround. Get all four wheels. Ninety seconds versus four hundred forty dollars.
What authoritative sources say
People also ask
Which Hertz airports have the AI scanners?
Hertz confirmed 10 US airports as of August 2025 and has signaled a target near 100 locations, but publishes no official list. Reported early sites include Atlanta, Charlotte, Houston, Newark, Phoenix, and Tampa. Treat any list you find online as unverified.
Can I refuse the AI scan?
Not really — the scanner sits in the return lane, so driving out is the scan. What you control is your own evidence. Film the car at pickup and at return.
Does a human review the AI's damage finding before I'm billed?
This is the core complaint. Hertz has been reported as issuing damage assessments without human review, which is what distinguishes it from competitors like Avis that describe their inspections as human-led.
How do I fight an AI damage charge?
Dispute in writing immediately and demand the before-and-after scan images that justify the charge. If that fails, dispute with your card issuer and file with your state attorney general's consumer protection office.
Is the scanner actually wrong, or is it just finding more?
Mostly the latter, and that's the uncomfortable part. It detects real damage a human would have missed. The dispute is about whether damage that small should trigger an automatic $440 bill — a pricing policy question, not a technology question.