When Canadian software engineer and YouTuber Anthony Sistilli built a DIY camera system to track police vehicles in Ontario, he set out to make a point about mass surveillance. When two officers turned up on his doorstep worried about their own personal privacy, he handed parents with tech-curious kids a perfect real-world case study in hardware, data, and digital ethics.
Canadian software engineer Anthony Sistilli built an automated license plate recognition (ALPR) system programmed exclusively to log police car movements, prompting Peel Regional Police officers to visit his home over privacy and safety concerns. The incident highlights the massive scale of public surveillance networks like Flock Safety, which operates over 120,000 cameras across 49 states, while giving parents a sharp hook to talk with teens about coding, privacy, and digital power dynamics. If your middle or high schooler wants to build intelligent camera projects, direct them toward tools like Scratch or ethical Python computer vision tutorials rather than live public tracking.
Sistilli decided to turn law enforcement's favorite surveillance tool back on itself. Using basic camera hardware, open-source software, and custom code, he built an Automatic License Plate Recognition camera designed to spot police cars, log their license plates, record timestamps, and map their locations around Brampton and Mississauga, Ontario. He documented the build on TikTok, describing it as a personal system to "Flock the Flockers."
About two weeks after testing it near a local shopping center, two officers from Peel Regional Police knocked on his door. Their primary concern was officer safety. One officer explicitly noted how unsettling it felt knowing someone could track when officers start their shifts, map their movements throughout the day, and potentially figure out where they live.
Sistilli pointed out the irony in real time: those exact concerns—being tracked without consent, having daily routines mapped, and facing potential data misuse—are identical to the concerns ordinary citizens raise about government and police surveillance every day.
Automatic license plate readers are not new, but their reach has grown dramatically. Companies like Flock Safety operate centralized surveillance networks across 49 states in the US, with over 120,000 cameras capturing billions of vehicle scans every month. Any police department on the network can search records across city and state lines, frequently without local public votes or explicit federal oversight.
While law enforcement uses ALPR systems to track stolen cars and locate suspects in serious crimes, the privacy trade-offs are real. Multiple investigations have documented police officers using Flock databases to stalk ex-partners or track individuals without a warrant. Similar camera networks have expanded into Canadian cities, where municipal governments spend millions on high-resolution ALPR setups with minimal public audit access.
When Sistilli pointed identical technology at police cruisers, the officers' reaction made the power dynamic obvious. Mass surveillance technology feels routine to the agency holding the database, but uncomfortable the moment the lens rotates.
If your kid is into coding, Raspberry Pi hardware projects, or cybersecurity channels on YouTube, a story like this grabs their attention immediately. Teenagers are drawn to projects that test boundaries or challenge authority, and building an automated machine-vision rig sounds infinitely cooler than working through a standard classroom coding exercise.
Instead of shutting down their interest or treating computer vision like a forbidden topic, use this story to show them how software engineering, legal boundaries, and ethics intersect.
- Separate the algorithm from the target. The underlying technology Sistilli used—optical character recognition and image classification—is basic computer vision. Learning how to train a model to recognize object shapes is a valuable technical skill. Choosing what or who to track is where ethics come into play.
- Understand public rights vs. tracking. Recording video in public space is legally protected under both Canadian law and the First Amendment in the US. Police confirmed Sistilli did not break the law, and no charges were laid. But aggregating continuous location data on specific human beings moves from public observation into personal tracking.
- Direct that energy toward ethical builds. If your teenager wants to build an AI camera, steer them toward contained home applications. An automated camera that texts them when the family dog gets on the couch or sorts Lego bricks by color uses the exact same software logic without invading anyone's privacy.
- On privacy double standards: "Why do you think the officers felt uncomfortable having their cars tracked, even though municipal cameras track every civilian car on the road?"
- On data retention: "Once location data is collected in a centralized database, who should own it, and how long should it stay on file?"
- On technical responsibility: "If you write code that automates data gathering, do you have a responsibility for how other people might use your tool?"
Q: Is it legal to build a camera that tracks police cars?
In both the US and Canada, filming or taking photos of police officers performing their duties in public spaces is legally protected. As long as a camera is set up on public property, does not obstruct law enforcement, and complies with local privacy and stalking laws, recording public vehicles on public roads is generally legal.
Q: What are Flock cameras and why are they controversial?
Flock Safety makes automated license plate readers used by thousands of police departments and neighborhood associations. The controversy stems from the massive scale of their centralized database, which allows law enforcement agencies to search vehicle movements across city and state lines without warrants or consistent oversight.
Q: How can I help my kid start a safe coding project with cameras?
Encourage them to explore open-source computer vision libraries like OpenCV using Python. Beginner projects like building an automated bird-feeder camera, an object sorter, or a motion-activated doorbell give kids hands-on experience with image processing without touching public surveillance or private data.
If your teenager is ready to turn their curiosity about hardware and code into actual projects, start with structured learning platforms. Try Code.org or explore Scratch for younger builders before diving into Python and computer vision. For age-based guidance on managing devices and technical independence, check out our digital guide for middle school and our digital guide for high school. To dive deeper into how software and AI impact education, explore our guide on tech in schools and our framework for kids using AI.