The US spent billions on border surveillance. Why can’t it catch people before they die?
MIT Technology Review James O'Donnell, Eileen Guo ● Covered by 3 sources
US border cameras missed a man who died 360 feet away. MIT says the virtual wall is catching fewer people than the money suggests.
Based on reporting by MIT Technology Review, James O'Donnell, Eileen Guo — read the original for the full story.
Summary, retelling and take written by AI under human oversight; images are AI-generated illustrations. How we work · Report an error
José Morales Bernal crossed into the United States on April 8, 2024, the day before his 32nd birthday. He was walking through the desert in southern New Mexico, inside the range of three surveillance towers built by Anduril for Customs and Border Protection. Those towers were supposed to spot him, alert agents, and get help moving fast if he needed it.
Instead, the first people to see him were workers at a nearby landfill. They spotted him again hours later, lying in the sand. Border Patrol showed up 45 minutes after the landfill workers called. Morales was dead, 360 feet from the nearest tower, with two more watching to the east and west. An autopsy later said he died of environmental exposure.
That single death sits inside a much larger failure. MIT Technology Review and Times of San Diego spent a year pulling together nearly 4,000 places where human remains were found and matching them against roughly 600 towers identified by the Electronic Frontier Foundation. Their finding is grim: more than 1,050 people died within range of border surveillance towers between 2015 and early 2026. Deaths showed up near nearly two-thirds of the towers studied, including modern autonomous towers from Anduril. Some towers could see only a sliver of the area they were advertised to cover. But many of the deaths were not hiding in blind spots.
The article also makes plain that this is not just a detection problem. It is a response problem too. In one case, two brothers were seen only after one waved down a helicopter; by then the younger man was dead. In another, a 24-year-old woman lay unnoticed for weeks near a tower that should have had a clear view. The government has spent 25 years and billions of dollars building what it calls a virtual wall, from older camera towers to blimps, drones, seismic sensors, and tunnel-sensing robots. Yet officials and former agents told the reporters that CBP does not formally investigate whether surveillance should have caught these people, or whether agents failed to reach them in time.
CBP says autonomous towers improve detection, situational awareness, and agent safety. Anduril says once a tower is delivered, CBP operates it. But the reporting argues that the basic promise of the system has not held up: too many people are still dying in places the government says it can see.
My take — AI-written commentary, not fact-checked reporting
The border tech pitch has been running on confidence for years, and the body count says it shouldn’t be. A camera that can watch a desert but not trigger a useful response is just expensive witness protection for failure. The real scandal here is not that the towers exist; it’s that Washington keeps buying more of them while acting surprised when the desert wins.
Read more about this at: MIT Technology Review
Related stories
AI models engage in ‘harmful activity directed at real people’, sparking fears safeguards not keeping up
CSET Georgetown · 1 month ago ·
34
AI powers citizen-led disaster relief from afar for Venezuela
Rest of World · 2 months ago ·
40
Texas Police Used AI to Write Report About Using Flock to Search for Woman Who Had Abortion
404 Media · 2 weeks ago ·
7