CBP bought 150 border scanners since 2020. A third sit in storage.

Customs and Border Protection has spent years and billions of dollars building out large-scale scanners meant to catch drugs and contraband moving through land ports of entry.▸ Two 2025 federal audits tested how much of that buildout actually works. A DHS Inspector General review⧉ found that 43 of the 150 large-scale scanners CBP purchased between 2020 and 2024 -- worth $96,017,586 -- were sitting in storage rather than deployed, and that 166 of the 361 large-scale scanners already installed at southwest border ports had gone inoperable at some point, for a combined 400,369 hours. A separate GAO review⧉ found CBP's own deployment plans currently exclude nine southwest border crossings -- including the two busiest land crossings in the country -- that together account for 38 percent of all southwest border passenger vehicle traffic.

Between 2020 and 2024, CBP purchased 150 fixed, large-scale Non-Intrusive Inspection (NII) systems from four contractors: 90 low-energy portals, 43 multi-energy portals, and 17 high-energy rail scanners, built to X-ray vehicles and rail cars for drugs and other contraband without requiring officers to search them by hand.▸ As of December 31, 2024, only 50 of the 150 -- a third -- were deployed and installed. Another 21 were deployed but not installed, 31 were still being built, and 43 were sitting in storage, with the deployment status of 5 more unknown.

The Inspector General's audit put a price on that idle inventory: the 43 systems in storage were worth $96,017,586.▸ The reasons varied. Two multi-energy portals sat for an average of 731 days because they interfered with existing radiation portal monitors. Three high-energy rail systems sat for an average of 611 days for lack of funding -- CBP had budgeted $4.4 million to prepare and install them, while the contractor doing the work estimated it would actually cost $47.4 million. One more rail system sat for 1,127 days because CBP couldn't resolve a land lease with the railroad that owned the site. CBP originally expected to finish deploying all 150 systems by the end of fiscal year 2025; it now estimates fiscal year 2029, and contractors have told the agency it will cost another $386 million to get there.

Storage isn't the only place scanners weren't scanning.▸ CBP had 361 large-scale NII systems installed at southwest border ports of entry, and the Inspector General reviewed 22,270 maintenance work orders filed on them between fiscal years 2019 and 2023. It found that 166 of those 361 systems -- 46 percent -- had gone inoperable at some point, for a combined 400,369 hours, with some repairs taking as long as 344 days. At one port, three low-energy portals that cost $32 million to procure and deploy went completely non-operational starting in June 2023; the port filed 135 work orders citing communication, image-quality, and X-ray problems before CBP gave up and decided to decommission and replace all three.

The Inspector General also found the data CBP uses to track its own scanners can't be trusted at face value.▸ In one instance, the agency's tracking platform recorded 134,410 vehicles scanned at a port of entry in December 2022 -- more than four times the 31,776 vehicles that actually crossed there that month, according to CBP's own separate crossing-count system. And of 508 logged equipment-interruption records from fiscal years 2019 through 2023, 32 percent were missing a category label describing what kind of problem occurred and 27 percent listed identical start and end times, meaning no one recorded how long the equipment was actually down. Thirteen different contracting officers and 14 different contracting officer's representatives rotated through the program's four maintenance contracts over that same six-year span.

The scanning buildout exists because Congress ordered it.▸ The Securing America's Ports Act requires CBP to scan 100 percent of vehicles entering at each land port of entry using large-scale NII by 2026, and the program has received over $2 billion since fiscal year 2019 -- $1,199 million for procurement, construction, and improvements, and $917 million for operations and support. CBP's own fiscal year 2021 Scanning Plan estimated the agency ultimately needs 434 large-scale systems, at an estimated cost of nearly $2 billion to acquire all of them. Individual system costs have climbed as CBP has tried to buy them: its estimate to install a single multi-energy portal rose from $1.3 million in 2020 to a $4.1 million average by 2023, and to as much as $9.7 million in an October 2024 estimate. One high-energy rail scanner planned for the Bridge of the Americas in El Paso was originally budgeted at about $4 million in 2021; by 2025, CBP said it needed $17.6 million more to finish the job.

GAO's review found something more specific than rising costs: CBP's current deployment plans simply don't reach some of the country's highest-traffic crossings.▸ Nine southwest border crossings -- accounting for 28,976,210 vehicle arrivals in fiscal year 2024, or 38 percent of all southwest border passenger vehicle traffic -- are excluded from the plans entirely. That list includes San Ysidro, California, the single busiest land crossing in the United States, and Otay Mesa, California, the second-busiest. Both were part of CBP's own 2020 deployment priorities before later plan updates dropped them, reportedly because there wasn't room to install large-scale equipment in their preprimary inspection areas.

That exclusion looks worse against CBP's own history at those exact crossings.▸ In fiscal year 2020, secondary-inspection scanning at San Ysidro, Otay Mesa, and the Nogales port of entry accounted for nearly 40 percent of all drug seizures at every land port of entry combined -- despite those scans covering less than 1 percent of total passenger vehicle traffic that year. GAO made two recommendations: that CBP clearly define and report on the two performance parameters it has never gotten around to measuring -- it currently tracks only operational availability, which itself hit only 51 percent of individual systems meeting CBP's own 95 percent target in fiscal year 2024 -- and that CBP figure out how to get its highest-traffic excluded crossings back into the deployment plan. DHS agreed with both.