The US Department of Homeland Security is seeking to develop a new mobile surveillance platform that fuses artificial intelligence, radar, high-powered cameras, and wireless networking into a single system, according to federal contracting records reviewed by WIRED. The technology would mount on 4x4 vehicles capable of reaching remote areas and transforming into rolling, autonomous observation towers, extending the reach of border surveillance far beyond its current fixed sites.

The proposed system surfaced Friday after US Customs and Border Protection quietly published a presolicitation notice for what it’s calling a “Modular Mobile Surveillance System,” or M2S2. The listing includes draft technical documents, data requirements, and design objectives.

DHS did not respond to a request for comment.

If M2S2 performs as described, border patrol agents could park their vehicles, raise a telescoping mast, and within minutes start detecting motion several miles away. The system would rely heavily on so-called computer vision, a kind of “artificial intelligence” that allows machines to interpret visual data frame by frame and detect shapes, heat signatures, and movement patterns. Such algorithms—previously developed for use in war drones—are trained on thousands if not millions of images to distinguish between people, animals, and vehicles.

The development of M2S2 comes amid the Trump administration’s sweeping crackdown on undocumented immigrants across the US. As part of this push, which has sparked widespread protests and condemnation for the brutal tactics used by immigration authorities,

Congress boosted DHS’s discretionary budget authority to roughly $65 billion. The GOP’s “One Big Beautiful Bill” allocates over $160 billion for immigration enforcement and border measures—most of it directed to DHS—with the funds scheduled to be distributed over multiple years. The administration has sought to increase DHS funding by roughly 65 percent, proposing the largest expansion in the agency’s history to fund new border enforcement, detention capacity, and immigration surveillance initiatives.

According to documents reviewed by WIRED, locations of objects targeted by the system would be pinpointed on digital maps within 250 feet of their true location (with a stretch goal of around 50 feet) and transmit that data across an app called TAK—a government-built tactical mapping platform developed by the US Defense Department to help troops coordinate movements and avoid friendly fire.

DHS envisions two modes of operation: one with an agent on site, and another where the trucks sit mostly unattended. In the latter case, the vehicles onboard AI would conduct the surveillance and send remote operators alerts when it detects activity. Missions are to be logged start to finish, with video, maps, and sensor data retained for a minimum of 15 days, locked against deletion “under any circumstances.”

The data collected by the units will be classified as Controlled Unclassified Information, or CUI, a designation introduced in the last decade to replace other labels such as “For Official Use Only,” referring to information that falls below the threshold for national security classification, but whose dissemination must be tightly controlled. (DHS considers any data that could reveal operational locations, network configurations, or personal information as restricted.) Even the program’s planning and testing documents will fall under this category.

Federal contractors are invited to review the proposal and submit feedback by late November. The agency said it expects to open formal bidding in early 2026, signaling that, while M2S2 is still in its early development phase, it is on a fast track for production.

Unlike earlier programs that relied on purpose-built vehicles, M2S2 is designed to be modular, its sensors, mast, and electronics capable of being removed and installed on other vehicles in less than a day. Ruggedized routers, switches, and antennas connect over cellular, radio, or satellite links, feeding imagery and tracking data to CBP command centers. With a fleet of such vehicles, each would act as a node in a wider surveillance mesh, capable of sharing its view with other units.

In a document outlining “optional” capabilities, bidders are encouraged to propose “additional equipment configurations” not specifically requested by the agency, which could connect the mobile units to other border patrol systems and surveillance towers.

Automation is a major requirement. The system must be capable of “autonomous detection and reporting” under any lighting or weather conditions. Its onboard AI must be capable of utilizing its computer vision capabilities fast enough for a remote operator to respond to an event in real time. This framework may later lend itself to cue other DHS assets, “including electronic warfare systems and kinetic systems (e.g., interceptor drones).”

Documents reviewed by WIRED outlining the project’s data-handling requirements reveal rules that show the system’s deep integration into CBP’s digital and cybersecurity framework. Every component, from cameras to routers, will carry their own unique identifiers. Networks must meet federal cybersecurity standards, with vulnerability scans and security reviews for each unit deployed.

M2S2 continues a lineage of CBP surveillance platforms that stretch back two decades. The “Mobile Surveillance Capability” trucks of the 2000s offered roving camera towers and radio links. The “Remote Video Surveillance Systems” of the period established the fixed towers now scattered across the border. The next decade introduced off-grid autonomous surveillance towers with AI-equipped masts powered by solar panels. While previous mobile units required manual control, autonomous surveillance systems remained strictly stationary. M2S2 by comparison can be driven into unmonitored terrain, operated with or without the presence of an agent, and tie its data back into CBP’s broader surveillance web.

Additionally, CBP wants the system to use open architecture so different manufacturers can integrate new tools without new code, reflecting a push to standardize its surveillance technologies. The objective is to avoid vendor lock-in while maintaining cybersecurity accreditation. According to presolicitation paperwork, CBP expects to award multiple blanket purchase agreements lasting up to 10 years. Early deployments would likely target areas lacking fixed tower coverage or sectors that require quick relocation after storms or migration surges.

The capabilities described by the documents pose a unique engineering challenge: fusing moving sensors, mobile networks, and AI analytics into a durable system able to survive heat, dust, and neglect. For CBP, it marks another step toward a surveillance network that is modular, shareable, and increasingly autonomous—observing more ground, for longer, with less need for agents in the field.