Orlando's Hottest Ride?
Someone Else's Car.

Hidden inside those almost 6,000 vehicle thefts between January 2021 and mid-2025, is a story about a viral social media trend, a manufacturing shortcut, and how quickly a city's crime landscape can be reshaped by a TikTok video.

One in six stolen vehicles in Orlando was a Kia or Hyundai.

In 2021, those two brands accounted for fewer than one in ten thefts. By 2025, that figure had climbed to nearly one in four — and remember: the summer 2023 data gap means those numbers are almost certainly an undercount.

The anti-theft vulnerability in certain Kia and Hyundai models was known to experienced car thieves for years. Then, in July 2022, a TikTok user in Milwaukee using the handle @robbierayyy posted a video showing how easily a Kia could be started using a screwdriver or the plug on the end of a USB cable — spreading the exploit to anyone with a smartphone.

The video, tagged with #kiaboyz, was up for just 13 days before TikTok took it down.

The damage, however, was done. Unlike many of their competitors, certain Hyundais (2015–2021) and Kias (2011–2021) with turn-key ignitions lacked basic anti-theft protections like electronic immobilizers, key fobs, and push-button ignitions. In an effort to cut costs, Hyundai Motor Group relied on traditional steel keys — leaving hundreds of thousands of vehicles without any meaningful electronic deterrent for theft.

The national numbers were staggering.

In Los Angeles, thefts of both brands jumped 85%. In Seattle, Kia thefts rose over 300% and Hyundai thefts over 500%. In the Twin Cities, re

Owners fought back in the courts. In 2021, a wave of class-action lawsuits was filed across the country, starting in Milwaukee with Marvin et al. v. Kia America, Inc., et al. Plaintiffs claimed the automakers knowingly cut costs by omitting engine immobilizers, failed to warn consumers about the heightened theft risk, and left owners financially exposed to theft, vandalism, higher insurance premiums, and diminished vehicle value.

Those lawsuits were eventually consolidated into one federal multidistrict litigation (MDL No. 3052), based in California. The case never made it to trial.

If you've driven around Central Florida, you've undoubtedly noticed our penchant for pickup trucks. We have a lot.

Conventional wisdom says pickups are the prime target for vehicle thieves, but in Orlando, the data tells a different story.

Sedans and coupes account for 41% of all stolen vehicles — nearly double the combined total of SUVs and pickups. The Kia Boyz effect largely explains this: both brands are heavily sedan-oriented, and the USB exploit specifically targeted lower-trim sedan models.

But this database contains far more than cars. Among the 5,929 vehicles stolen in Orlando over five years: 127 motorcycles, 144 utility and contractor trailers, 49 commercial trucks, 22 pieces of lawn and construction equipment, 9 buses, 6 RVs — and 5 manufactured homes.

Orlando's overall vehicle recovery rate — 74.8% — is surprisingly high. But that average conceals a dramatic split. Passenger cars recover at over 80%, largely because thieves tend to abandon them nearby once they run out of gas or they lose interest. Trailers recover at 29%. Lawn and utility equipment at just 18%.

The most sobering statistic: five manufactured homes were stolen in Orlando over five years. Only one was recovered. Someone stole four houses and kept them.

The data records when thefts were reported — not when they occurred. That distinction matters: the Monday morning spike almost certainly reflects cars stolen over the weekend that owners discover when they leave for work. The actual theft likely happened on a Friday or Saturday night.

The overnight hours between midnight and 5am show the fewest reports — not because fewer thefts happen then, but because fewer people are awake to immediately discover them. The real theft-time distribution is almost certainly heavier overnight.

The theft hotspot map reveals two distinct stories. The large purple cluster near Orlando International Airport almost certainly represents a data artifact: rental cars are reported stolen at the airport for a variety of reasons including someone accessing the wrong car during a self-service format (a vehicle being reported as "stolen" only to pop up later) or a customer making a very late rental car return.

Orlando International Airport hosts a lot of visitors who are unfamiliar with car rentals. Nearly 977 records, about 12% of the database, fall into this category.

Strip those records out lot of foreign visitors and people who are unfamiliar with car rentals. Nearly 977 records, about 12% of the database, fall into this category.

Strip that out and a clear geographic pattern emerges: thefts concentrate heavily on Orlando's west side, along the Kirkman Road corridor, around Millenia Mall, and stretching north through Pine Hills along the Orange Blossom Trail.

If you own a Hyundai or Kia, your next steps depend on which bucket you fall into. Critics note that $4.5 million is not nearly enough to cover actual victim costs when you factor in towing, deductibles, higher insurance premiums, and lower resale values. And the qualifying window for the $9M settlement is narrow.

To qualify for the $9M state settlement, the theft must have occurred after April 29, 2025 (and before the vehicle received an ignition-cylinder protector), or no later than March 31, 2027, whichever comes first. At the time of the theft, your vehicle must have had the software upgrade or had an appointment scheduled. Eligible owners can claim up to $4,500 for total loss, up to $2,250 for partial loss, or up to $375 for damage from an attempted theft.

One critical note: once the settlement money is gone, it's gone. Hyundai and Kia will not add more once funds are exhausted.

This analysis began with a public records request submitted by WKMG-TV investigative reporter Mike DeForest to the Orlando Police Department in early 2025. What came back was a raw spreadsheet of 11,447 rows covering vehicle theft and recovery records from January 2021 through mid-2025 — a dataset that had never been publicly analyzed at this level of detail.

The cleaning, categorization, and analysis work was done by Donovan Myrie, Ph.D., using Claude (Anthropic) as a data analysis partner. The full methodology is documented below. The cleaned dataset is available for download.

All analysis was conducted in Python using pandas and openpyxl. Charts were built with Chart.js. The interactive map uses Leaflet.js with OpenStreetMap/CARTO tiles. The full geocoded address dataset covering all 2,763 unique theft locations is available to journalists and researchers upon request.