ICE Uses Vendor Loopholes, Mobile Apps to Expand U.S. Surveillance Web

How ICE's Surveillance System Works — Wendover Productions

Wendover Productions

The federal immigration agency is leveraging private tech contracts and faulty biometric tools to bypass local sanctuary laws, laying the groundwork for a broader domestic surveillance state that targets both immigrants and political protestors.

AI-assisted summary of How ICE's Surveillance System Works — original source. Check the original for full context.

Source published:

Detailed notes

## Background Published on YouTube on March 24, 2026, by the channel Wendover Productions, this video breaks down how the Immigration and Customs Enforcement agency utilizes digital surveillance. The video description provides an exclusive NordVPN promotional link with four extra months, sponsorship and production contact emails, and credits Sam Denby for writing and Alexander Williard for editing.

## TL;DR The video reveals how Immigration and Customs Enforcement (ICE) circumvents legal and state restrictions to access a nationwide data-sharing network built by private surveillance startups like Flock Safety. Additionally, ICE deploys custom-built biometric facial recognition apps like Mobile Fortify in the field—populating its databases through airport exit checks—and utilizes cell site simulators known as stingrays to track phones without warrants. Driven by a newly approved $75 billion budget, these highly automated, error-prone digital tools are increasingly used as a pretext for unlawful detentions and to build databases monitoring political protesters.

## Atmosphere & style The presentation carries the characteristically analytical, investigative, and deeply clinical tone typical of a high-production journalistic documentary. It blends an overarching sense of systemic dread with sharp, objective alarm as it systematically peels back the layers of a hidden, corporate-backed governmental apparatus. Watching it feels like a sobering dive into algorithmic overreach, shifting from localized municipal loopholes to a sweeping critique of a creeping American surveillance state that increasingly normalizes the erosion of constitutional privacy.

## Key takeaways - [00:00](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=0s) ICE accesses AI-enabled license plate reader data in non-cooperative municipalities by leveraging private networks operated by venture-backed startups like Flock Safety. - [01:41](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=101s) Automated data-sharing agreements allow outside law enforcement agencies cooperating with ICE to easily query and access localized surveillance data repositories. - [01:41](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=101s) The lack of standardized justification requirements for data searches leaves local authorities entirely unaware of how outside agencies utilize their surveillance networks. - [03:29](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=209s) Flock Safety data has been repeatedly abused by rogue law enforcement officers for unauthorized personal surveillance and tracking. - [05:00](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=300s) Even when municipal access is restricted, ICE can bypass local restrictions by tapping into unregulated private commercial networks operated by retail corporations. - [06:29](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=389s) ICE and Border Patrol use a proprietary mobile facial recognition app called Mobile Fortify to instantly extract biographical and legal data from biometric databases. - [06:29](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=389s) The biometric data powering field recognition tools is siphoned heavily from expanded, long-term exit checks conducted on non-citizens at U.S. international airports. - [09:18](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=558s) Mobile Fortify displays high technical inaccuracy in real-world field environments, leading to misidentifications that are still used as immediate pretexts for arrest. - [09:18](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=558s) Federal agents utilize field facial recognition app data to build long-term 'databases of dissent' by tracking and identifying political protesters. - [13:59](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=839s) ICE deploys highly invasive cell site simulators known as stingrays to intercept IMSI numbers from mobile devices within a targeted area. - [14:51](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=891s) Despite internal policies requiring targeted judicial warrants, official oversight confirms ICE routinely uses stingrays for warrantless mass surveillance. - [15:46](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=946s) A recently approved $75 billion funding surge from Congress will nearly triple ICE's annual budget, accelerating the automation and expansion of its digital surveillance suite. - [16:29](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=989s) ICE tolerates flawed data and high system error rates because the technology serves its operational goal of providing a convenient pretext for unlawful, post-detention investigations. - [17:14](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=1034s) The rapid expansion of ICE's methods threatens the Fourth Amendment by normalizing a government-wide mass surveillance state that logs the locations and associations of all citizens.

## Detailed walkthrough ### [00:00](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=0s) Backdoor Surveillance in Glenwood Springs

The federal Immigration and Customs Enforcement agency (ICE) has found a convoluted way to access data from AI-enabled license plate reading surveillance cameras. This is occurring even in municipalities like Glenwood Springs, Colorado, which are deeply unsupportive of ICE's mission. Glenwood Springs voted two to one for Kamala Harris in the 2024 election and features a large immigrant population where roughly one-third of households are Spanish-speaking. Furthermore, Colorado voters supported a statewide blue majority law explicitly prohibiting local police from cooperating or sharing data with federal immigration authorities.

The loophole exists because while the local police department operates the camera, they do not own it. The system is owned by Flock Safety, a fast-growing, venture capital-backed startup. Glenwood Springs pays an annual fee of $37,500 to access the data. The software suite allows local police to look back in time to see exactly when and where a vehicle passed through town, or add a plate to a "hot list" to trigger an arrest alert. Internal logs reveal the department uses these cameras nearly every day to investigate everything from assault and stalking to sex offenses. Glenwood PD reports the system has been instrumental in notable successes, including arresting a California gang member in a stolen vehicle, a wanted felon suspected of elder abuse, and an Oregon pedophile who lured a child to the town.

* **Glenwood Springs Demographics:** Two to one vote for Kamala Harris in 2024; one-third of residents speak Spanish at home. * **Annual System Cost:** $37,500 paid to Flock Safety. * **Geographic Advantage:** Interstate 70 passes directly through Glenwood Springs, channeling thousands of long- and short-distance commuters through the cameras daily.

---

[01:41](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=101s) Data Sharing Agreements and Political Dissent

Because Interstate 70 serves as a major escape route for fleeing suspects, law enforcement agencies nationwide desire access to Glenwood's camera network. Glenwood Springs opted into automated data-sharing agreements, allowing outside departments to search their repository at the press of a button in exchange for reciprocal access. However, this network is exactly "how ICE got in." In January 2025 alone, outside police departments queried Glenwood's data 516,892 times. A portion of these external agencies openly cooperate with ICE. Internal logs explicitly acknowledge that Homeland Security Investigations (HSI), the investigative arm of ICE, has conducted searches using Glenwood's data.

The surveillance network also shows potential for tracking political dissent. Following the nationwide "No Kings Day" protest against the Trump administration, October logs revealed multiple searches justified simply with the logged reasoning: "protest." While crimes can occur during peaceful demonstrations, logging a protest itself as the justification demonstrates a willingness to use Flock data in illegitimate ways. This risk is amplified because there is no standardized threshold requiring officers to justify their searches. Thousands of times per month, outside agencies log ambiguous search reasons like "investigation," "TBD," "test," or simply the number "0," leaving Colorado authorities entirely in the dark.

* **External Access Count:** 516,892 data accesses by outside departments in January 2025 alone. * **Ambiguous Log Entry Examples:** "investigation", "TBD", "test", "0".

---

[03:29](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=209s) Rogue Actors and Contractual Loopholes

Ample precedent exists regarding rogue law enforcement officers using Flock data for nefarious personal reasons. Because tracking is so spotty, rogue actors easily fly under the radar, and those who are caught typically face internal disciplinary action rather than legal prosecution. * A Kansas police chief used Flock data to stalk his ex-girlfriend 228 times. * A North Charleston, South Carolina officer used the system to stalk his wife. * An officer in Miami tracked his girlfriend, and an officer in Menasha, Wisconsin stalked his ex-girlfriend.

When Glenwood Springs residents discovered ICE's backdoor access, they pressured the city council. The police department initially responded by cutting off system access to all law enforcement agencies outside of Colorado. However, the *Aspen Daily News* reported that two police departments within Colorado illegally bypassed state law to share data with the federal agency anyway. In response, Glenwood shut off outside access entirely. Despite this, the city attorney noted that their contract with Flock meant they "can't completely control what Flock does with data from Glenwood's system." This concern is validated by Denver, located three hours away, which terminated its Flock contract after discovering the city's data was shared with federal immigration authorities against their knowledge as part of a secret pilot program.

---

[05:00](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=300s) Private Backdoors and National Concerns

Even if a city completely walls off its police cameras, ICE can still exploit private commercial networks. At the entrance of the local Lowe's Home Improvement parking lot in Glenwood Springs, private Flock Safety cameras log every single car entering the lot. Unlike public law enforcement, Lowe's does not publish its logs or disclose who has access to its database. Thousands of these corporate-operated cameras exist nationwide. Because these private entities do not answer to public constituents, they effectively "act as a backdoor for ICE to surveil the movements of Americans even in non-cooperative states and municipalities." Recent polling reveals a majority of Americans distrust ICE, making this commercial data pipeline a growing concern. Because most U.S. police departments do not face the public pressure Glenwood did, they continue to share data freely. Any database is only as secure as its weakest link, giving ICE open access to a nationwide surveillance web simply by finding one friendly local department.

---

[06:29](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=389s) Mobile Fortify and Airport Biometrics

ICE does not always rely on roundabout methods; many tools are custom-built for the agency. In videos documenting tense or violent immigration arrests, agents are increasingly seen holding smartphones directly up to people's faces. They are using a proprietary facial recognition app called "Mobile Fortify" (and a closely related Border Patrol variant). In New York, an agent was documented using the app to identify an individual waiting for an immigration hearing at a federal courthouse. A successful scan instantly pulls up the target's name, date of birth, nationality, immigration status, and any prior deportation orders, establishing immediate probable cause for an arrest.

To populate this database, ICE siphons biometric data from Customs and Border Protection (CBP). CBP has drastically expanded biometric gathering at U.S. airports. Previously, the U.S. lacked formal exit immigration controls, relying instead on flight manifests provided by airlines. Today, biometric gates scan travelers' faces and cross-reference them against passport photos. By late 2025, DHS announced these exit checks would become universal at all international airports. While photos of U.S. citizens are reportedly deleted within 12 hours, non-citizens' photos are retained for up to 75 years, flowing directly into the Mobile Fortify ecosystem alongside data from ICE's case management system and the FBI's National Crime Information Center.

> "ICE does not provide the opportunity for individuals to decline or consent to the collection and use of biometric data/photograph collection." — *Internal ICE Document*

---

[09:18](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=558s) Faulty Field Logic and Databases of Dissent

In practice, Mobile Fortify is plagued by technical inaccuracies. During an October 2025 ICE raid in the agricultural town of Woodburn, Oregon, a legal deposition revealed the app's severe field limitations. An officer scanned a detained woman's face and admitted, "it came up to a very like a similar person. I wasn't sure if it was her or not." The system suggested the name "Maria." When the woman did not respond, the officer scanned her again, and the app generated an entirely different identity match.

When asked in court to state his legal probable cause for the arrest, the officer testified that he believed she was undocumented because she spoke Spanish to interact with others and because she "became a possible match on my facial recognition system." Though courts have ruled that language alone cannot establish probable cause, this officer relied entirely on an app he knew had misidentified the woman twice. In real-world environments—characterized by poor lighting, shaky phone cameras, and moving subjects—facial recognition routinely struggles.

Beyond identification, Mobile Fortify is being used as a tool to monitor political protesters. In cities like Minneapolis, federal agents have been seen holding up phones to scan anti-ICE protesters. Because these crowds are overwhelmingly white and English-speaking, agents have no reason to suspect irregular immigration status. Instead, they are exploiting a policy mismatch: while airport data is highly regulated, field photos captured on Mobile Fortify are retained for up to 15 years, even if the person is a confirmed U.S. citizen.

The app generates an "encounter log" tying a person's photo and identity to a specific time and location. In Minnesota, a woman serving as a legal observer walked behind DHS vehicles to document their actions. An officer blocked her path, addressed her by her full name despite her never giving it, warned her they had facial recognition technology, and threatened her with arrest. Three days later, the TSA revoked her 14-year-old PreCheck status. Furthermore, when an agent scans someone, Mobile Fortify allows them to see if other agents have previously logged that individual at other protests, effectively turning the platform into a "database of dissent" used to target specific activists.

---

[13:59](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=839s) Cell Phone Surveillance Via Stingrays

As the surveillance tools get more sophisticated, their operational details become more guarded. ICE is known to possess "stingrays"—highly sophisticated cell phone surveillance systems typically built into vehicles costing half a million dollars.

Stingrays operate by spoofing legitimate cellular towers. Because mobile phones are hardwired to constantly seek out and connect to the strongest available local signal, any phone in the vicinity of a stingray will automatically drop its carrier connection and connect to the fake police signal instead.

Once a connection is established, the stingray intercepts the phone's International Mobile Subscriber Identity (IMSI) number, a unique numerical identifier tied directly to the user's cellular carrier account. Armed with this IMSI number, ICE can issue a subpoena, court order, or search warrant to the carrier to unmask the target's true identity. Because stingrays are incredibly invasive and capture data from all surrounding innocent bystanders, internal ICE policies dictate that they may only be deployed for targeted investigations backed by a judicial warrant.

[14:51](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=891s) Warrantless Stingrays and Mass Protest Surveillance

While cell site simulators, commonly known as stingrays, can be legitimately useful for determining whether a targeted individual is at home or at work, Immigration and Customs Enforcement (ICE) has a long history of ignoring its own policies and using them without a warrant. This practice is so pervasive that it prompted an entire Inspector General report dedicated to the issue. Furthermore, circumstantial evidence suggests that ICE has flagrantly broken policy by deploying stingrays in non-targeted mass surveillance operations, capturing the IMSI numbers of anyone who happens to be in a target area.

A stark example of this occurred during an anti-ICE protest on July 4, 2025, in Seattle, Washington. An analysis of cellular signals at the demonstration revealed anomalies consistent with the use of an IMSI catcher. This indicates that the agency may have been using these devices to harvest a comprehensive log of all mobile devices—and therefore every individual—present at the protest.

- **Targeted vs. Mass Surveillance:** Stingrays are designed to locate specific individuals, but ICE has used them broadly to sweep up data on entire crowds. - **Seattle Protest:** Cellular signal anomalies detected on July 4, 2025, pointed directly to IMSI catcher deployment by the agency. - **Official Oversight:** The misuse of this technology is documented in an official Inspector General report.

---

[15:46](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=946s) The Automated Deportation Pipeline and ICE's Ballooning Budget

License plate readers, mobile fortify systems, and stingrays represent merely the tip of the iceberg. ICE currently commands a massive software suite of digital surveillance tools that increasingly automate their investigative operations. This automation allows more field personnel to remain on the streets, actively rounding up more individuals for deportation. Because ICE rarely volunteers information regarding its surveillance capabilities, the public is left in the dark, relying on leaks, court documents, mandatory disclosures, and direct observation to piece together what tools are being used.

The scope of this surveillance is poised to expand drastically. Congress recently approved an additional $75 billion in funding for ICE, nearly tripling the agency's annual budget over the next few years. This massive financial influx ensures that the depth and breadth of its monitoring systems will accelerate rapidly, posing an escalating threat to immigrant rights.

- **Automation:** Digital surveillance suites automate investigations, shifting personnel from desks to active street roundups. - **Information Control:** Public knowledge of ICE's tech relies strictly on leaks, court disclosures, and observation. - **Budget Surge:** Congress approved $75 billion in additional funding, nearly tripling ICE's annual operating budget.

---

[16:29](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=989s) Pretextual Arrests and the Erosion of Constitutional Rights

Many of the digital tools deployed by ICE suffer from massive error rates and are fundamentally ineffective. However, the agency has little motivation to improve them because, for their specific objectives, the tools work perfectly fine. They serve as a convenient pretext for unlawful arrests, allowing agents to detain individuals who lack legal status without legitimate probable cause. Because establishing true probable cause is difficult, the agency finds it much easier to simply ignore constitutional rights, detain people first, and investigate their residency status later.

This operational model persists because no one currently in power is in the business of challenging ICE's lawfulness. Furthermore, many Americans actively choose to ignore the constitutional and human rights violations of immigrants, provided it hastens the administration's progress toward "their goal of a more homogeneous America."

- **Pretextual Detentions:** High error rates are tolerated because flawed data is still used as a pretext to bypass standard probable cause requirements. - **The Process:** ICE frequently detains individuals first and investigates their legal residency status afterward. - **Political and Social Backing:** A lack of congressional or executive pushback, combined with a public desire for a homogeneous society, insulates the agency from accountability.

---

[17:14](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=1034s) The Creep of the American Surveillance State

Even for citizens who support ICE's ultimate goals, the agency's methods should remain a serious source of concern. ICE has played a pivotal role in a massive, government-wide creep of mass surveillance that has accelerated across multiple presidential administrations. Historically, the United States public has been uniquely wary of domestic government surveillance—polling consistently shows Americans care far more about this issue than the average European, traditionally preferring a compromise in security to ensure increased privacy.

ICE is directly challenging the Fourth Amendment's protections against unreasonable search and seizure by effectively searching everyone without a warrant in the hopes of finding useful information. While some might justify these "fast and loose methods" because they make the removal of immigrants more efficient, the long-term consequence is the normalization of an America where those in power know where you are, who you know, and what you are doing at any given time. This power is already being abused against immigrants and protestors, but it can easily be directed toward anyone in the future.

> "America is becoming a surveillance state, and too many aren't concerned because the victims so far are primarily those they've been told are the enemy."

---

[18:49](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=1129s) Mitigating Surveillance and the Practical Utilities of a VPN

While many forms of digital surveillance are completely unavoidable, there are ways to protect your data. For instance, internet service providers (ISPs) routinely track and store browsing history, which can be handed over to the US government via a proper subpoena or warrant. Utilizing a service like NordVPN addresses this by establishing an encrypted tunnel between the user and a secure server. Unlike traditional ISPs, NordVPN enforces a strict no-logs policy, meaning they retain no usage data to hand over to authorities even if legally compelled—a claim independently audited and verified by both PwC and Deloitte.

Beyond dodging government surveillance, a VPN offers several day-to-day practical applications. It allows users to bypass government firewalls and circumvent geo-blocking on streaming websites by virtually redirecting traffic through servers in various countries. Additionally, features like Threat Protection Pro enhance general internet security by automatically blocking dangerous ads, pop-ups, and malicious websites, drastically reducing the risk of malware.

- **Audited Security:** NordVPN’s zero-logs policy has been verified through independent reviews by PwC and Deloitte. - **Censorship and Geo-blocking:** Traffic can be virtually routed to access an open internet or unlock region-restricted streaming content. - **Threat Protection Pro:** Built-in security tools actively block malicious ads and reduce malware risks. - **Viewer Offer:** NordVPN is offering an exclusive deal for Wendover viewers that includes 4 extra months at [nordvpn.com/wendover](https://nordvpn.com/wendover), backed by a risk-free 30-day money-back guarantee.

## Notable quotes > [06:29](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=389s) ICE does not provide the opportunity for individuals to decline or consent to the collection and use of biometric data/photograph collection. — Internal ICE Document

> [09:18](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=558s) it came up to a very like a similar person. I wasn't sure if it was her or not. — ICE Officer

> [09:18](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=558s) became a possible match on my facial recognition system. — ICE Officer

> [17:14](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=1034s) America is becoming a surveillance state, and too many aren't concerned because the victims so far are primarily those they've been told are the enemy.

## Mentioned Immigration and Customs Enforcement, ICE, Wendover Productions, NordVPN, Sam Denby, Alexander Williard, Sara Stoltman, Gabriel Ferreras, Austin Glass, Kate Ermolenkko, Manni Simon, Dony Bullen, Simon Buckmaster, Flock Safety, Kamala Harris, Glenwood Springs Police Department, Interstate 70, Homeland Security Investigations, HSI, Trump administration, Aspen Daily News, Lowe's Home Improvement, Mobile Fortify, Border Patrol, Customs and Border Protection, CBP, Department of Homeland Security, DHS, FBI, National Crime Information Center, TSA, PreCheck, PwC, Deloitte, Threat Protection Pro

[Source](https://www.youtube.com/watch?v=aRUhgrqnpGY)

中文简报

ICE利用供应商漏洞与移动应用扩大全美监控网络 该联邦移民机构正利用私营技术合同和存在缺陷的生物识别工具绕过地方庇护法,为建立一个针对移民和政治抗议者的更广泛的国内监控国家奠定基础。 该视频揭示了美国移民与海关执法局(ICE)如何规避法律和州层面的限制,获取由Flock Safety等私营监控初创公司建立的全国性数据共享网络的访问权。此外,ICE在实地部署了名为Mobile Fortify等定制生物识别面部识别应用程序(通过机场出境检查扩充其数据库),并在没有搜查令的情况下使用被称为“黄貂鱼”(stingrays)的基站模拟器追踪手机。在新获批的750亿美元预算推动下,这些高度自动化且容易出错的数字工具正日益被用作非法拘留的借口,并用于建立监控政治抗议者的数据库。 ICE通过利用Flock Safety等风投支持的初创公司运营的私营网络,获取不合作城市的AI车牌识别数据。 (0:00) 自动化数据共享协议使与ICE合作的外部执法机构能够轻松查询和访问本地化的监控数据库。 (1:41) 数据检索缺乏标准化的正当理由要求,导致地方当局完全不知晓外部机构如何使用其监控网络。 (1:41) Flock Safety的数据曾多次被违规执法人员滥用,用于未经授权的个人监控与追踪。 (3:29) 即使市政访问受限,ICE也可以通过接入零售企业运营的不受监管的私营商业网络来绕过地方限制。 (5:00)

Reading summary

Wendover Productions reports that Immigration and Customs Enforcement (ICE) has constructed an expansive surveillance apparatus by exploiting commercial data-sharing loopholes, biometric databases, and warrantless cellular tracking. The program argues that despite high error rates, these automated systems provide pretexts for detentions while eroding constitutional protections against unreasonable search and seizure. Commercial Loopholes and License Plate Networks (0:00) ICE routinely bypasses state and municipal sanctuary policies by tapping into automated license plate recognition networks operated by private contractors like Flock Safety. In Glenwood Springs, Colorado—a municipality that pays $37,500 annually to use Flock's cameras and is bound by state laws barring cooperation with federal immigration authorities—outside police departments accessed local camera data 516,892 times in January 2025 alone. Internal logs confirmed Homeland Security Investigations conducted queries through these reciprocal networks. External agencies frequently logged uninformative justifications such as 'investigation', 'test', '0', or 'protest', preventing local oversight. Municipalities face contractual limitations in restricting data dissemination. Glenwood Springs' city attorney acknowledged an inability to fully dictate Flock's data practices, a dilemma highlighted when Denver previously terminated its contract after learning its data was funneled into a secret federal immigration pilot. Furthermore, ICE can circumvent municipal barriers entirely by accessing Flock cameras deployed by private commercial entities, such as retail parking lots, which operate without public disclosure requirements. Biometric Data Siphoning and Flawed Facial Recognition (6:29) ICE agents increasingly utilize 'Mobile Fortify,' a proprietary mobile facial recognition application populated with biometric records drawn from Customs and Border Protection (CBP) and the FBI. Under expanding CBP airport programs, non-citizen biometric photos are retained for up to 75 years without an option to decline or consent, feeding directly into ICE's enforcement platforms. In the field, agents use the app to instantly surface identity, nationality, and deportation histories. Legal depositions reveal that the software frequently falters under field conditions. During an October 2025 enforcement action in Woodburn, Oregon, an agent detained a woman after Mobile Fortify generated two conflicting identity matches, with the officer testifying that the inconclusive scan and the subject's Spanish language formed his probable cause. Additionally, federal agents have deployed Mobile Fortify to scan English-speaking demonstrators at anti-ICE rallies in Minneapolis. Field encounter photos are retained for up to 15 years regardless of citizenship status, enabling the agency to build cross-referenced records tracking political activists. Warrantless Cellular Interception and Protest Monitoring (13:59) ICE deploys vehicle-mounted cell-site simulators, commonly known as stingrays, which spoof legitimate cell towers to force nearby mobile phones to connect to fake signals. This allows agents to intercept International Mobile Subscriber Identity (IMSI) numbers and unmask user identities via carrier subpoenas. Although internal policies mandate judicial warrants due to the indiscriminate collection of bystander data, a Department of Homeland Security Inspector General report documented systemic warrantless deployments by the agency. Circumstantial evidence indicates ICE has utilized stingrays for non-targeted mass monitoring rather than localized suspect tracking. During a July 4, 2025 demonstration in Seattle, Washington, cellular signal anomalies pointed to the deployment of an IMSI catcher, suggesting the agency harvested mobile device identities across the assembled crowd without individualized suspicion. Automated Enforcement, Pretextual Detentions, and Constitutional Erosion (15:46) Congress recently approved an additional $75 billion in funding for ICE, nearly tripling its annual operating budget and accelerating the deployment of automated digital investigative suites. Wendover argues that ICE tolerates inaccurate and defective surveillance tools because they still function effectively as pretexts for unlawful detentions. Rather than establishing legitimate probable cause beforehand, agents use flawed technological flags to detain individuals first and investigate their residency status subsequently. This operational model persists due to a lack of institutional pushback from the executive and legislative branches, alongside public tolerance of immigrant rights violations. Wendover contends that by normalizing universal, warrantless data collection across commercial, biometric, and cellular domains, ICE is advancing a domestic surveillance infrastructure that threatens broader Fourth Amendment protections and can readily be turned against the wider public. Wendover concludes that ICE's growing surveillance ecosystem represents an accelerating drift toward a domestic surveillance state. While justified by proponents as an efficient mechanism for deportations, the unchecked combination of commercial data brokering, error-prone biometric matching, and warrantless cellular interception establishes operational precedents that undermine constitutional safeguards for all Americans.

精炼阅读摘要

Wendover Productions 报道指出,美国移民及海关执法局(ICE)正通过商业车牌识别系统、移动面部识别应用与车载伪基站等技术,系统性规避地方保护性法律与宪法第四修正案限制,构建起一个缺乏外部问责且不断扩张的国内监控体系。 车牌识别共享与商业监控后门 (0:00) 即使在明确立法禁止地方警察协助联邦移民执法的科罗拉多州,ICE 仍能间接获取车牌监控数据。以 Glenwood Springs 为例,该市每年向风投支持的初创企业 Flock Safety 支付 37,500 美元使用其 AI 车牌识别网络。由于 Interstate 70 贯穿该地,各警务机构建立了互惠数据共享机制;仅 2025 年 1 月,外部机构对该市数据进行了 516,892 次查询,其中与 ICE 合作的机构以及国土安全调查局(HSI)均记录在案。外部人员查询时常使用“test”、“0”或“investigation”等含糊理由,甚至在全美“No Kings Day”抗议后直接以“protest”为由调取行车记录。 由于监管缺失,系统曾多次被个别警员滥用于私人跟踪,例如堪萨斯州一名警察局长曾查询前女友行踪达 228 次。面对公众抗议,Glenwood 虽切断了外州甚至外部直接访问权限,但市律师承认合同上无法完全掌控 Flock 对数据的处置,丹佛此前就因数据被秘密纳入联邦试点项目而与 Flock 解约。此外,Lowe's 等大型零售商在停车场设立的私营 Flock 摄像头不对外公开日志,成为不受公众监督的商业数据后门,使 ICE 能在全美自由调用私营监控网络。 移动面部识别与生物数据外溢 (6:29) 在现场执法中,ICE 特工广泛使用名为 Mobile Fortify 的定制面部识别应用。特工直接将智能手机对准被检查者面部,即可调取其姓名、出生日期、国籍、移民状态及遣返记录以确立逮捕依据。该应用的数据源依托海关及边境保护局(CBP)在机场扩建的生物识别闸机。国土安全部(DHS)宣布到 2025 年底将在全美所有国际机场普及离境面部扫描;非美国公民的面部图像数据将被保留长达 75 年,并同步给联邦调查局(FBI)及 ICE 数据库。ICE 内部文件明确指出,当事人无权拒绝或同意生物特征数据的采集。 实际应用中,该系统在移动环境与暗光条件下错误频发。2025 年 10 月俄勒冈州 Woodburn 的一次农场突袭诉讼证实,一名特工在系统先后比对出两个不同姓名且已知存在误认的情况下,仍以当事人讲西班牙语和算法提供“疑似匹配”为由实施逮捕。与此同时,特工被发现在明尼阿波利斯等地的反 ICE 示威中扫描以白人为主的抗议人群;现场拍摄的照片最长可保留 15 年,平台生成的“接触日志”能够关联个体的出现时间与地点,实际上沦为追踪特定异见人士与法律观察员的政治监控工具。 车载伪基站与无令状抗议监控 (13:59) 除了光学识别,ICE 还装备了单车改装造价约 50 万美元的“Stingray”(黄貂鱼)蜂窝基站模拟器。该设备通过伪造强信号基站迫使周围所有手机断开正常网络并接入伪基站,从而拦截用户的国际移动用户识别码(IMSI)。特工随后可借此向电信运营商索取机主的真实身份。 尽管 ICE 内部章程规定 Stingray 具有极强侵入性、必须在获得司法搜查令后方可针对特定目标部署,但内部监察长报告披露,该机构长期存在无令状违规使用的情况。环境证据表明,ICE 曾在非定向群体场景中使用该设备;例如在 2025 年 7 月 4 日华盛顿州西雅图的反 ICE 抗议集会期间,现场检测到的蜂窝信号异常高度符合 IMSI 拦截器的特征,显示该技术可能已被用于大范围采集抗议人群的手机身份标识。 预算倍增与宪法第四修正案侵蚀 (15:46) 美国国会近期批准为 ICE 增拨 750 亿美元,使其年度运营预算在未来数年内近乎翻三倍,这将显著加速其自动化监控与驱逐体系的部署。数字调查工具的大范围应用使文职特工能将更多精力投入街头抓捕,而公众仅能通过法庭文件、泄密和偶发观察拼凑其真实能力。 由于缺乏国会与行政分支的实质性问责,加上社会部分群体对移民群体的漠视,ICE 往往利用高误报率的算法结果作为实施“借口性逮捕”(pretextual arrests)的由头,普遍采取“先拘留、后调查”的模式以绕过合理的逮捕依据。节目指出,这种无差别的数据筛查正从根本上削弱宪法第四修正案对免受无理搜查的保障,将针对边缘群体的执法特权逐步常态化为覆盖全体居民的国家级监控网络。 本节目指出,ICE 正通过公私合作协议与先进技术工具规避法定限制,其实际监控范畴已越过移民事务延伸至政治集会与公共生活。尽管这些手段被辩护为提高执法效率的工具,但对其合法性审查与技术缺陷的纵容,正在全美范围内加速一种不受约束的泛在化监控常态。

Detailed notes — zh

## Background 该视频由 YouTube 频道 Wendover Productions 于 2026 年 3 月 24 日发布。视频简介列出了主要制作人员名单,提供了赞助商 NordVPN 的优惠信息,并附上了关于 Flock Safety 车牌监控案例、警员系统滥用丑闻、公众对 ICE 信任度调查以及 Mobile Fortify 隐私影响评估文件的多项参考资料。 ## TL;DR 视频系统性揭示了美国移民和海关执法局(ICE)如何编织一张无孔不入的全国监控天网。通过接入 Flock Safety 的商业车牌识别网络与零售商探头,联邦特工得以绕过地方庇护法律获取车辆轨迹;同时,前线执法人员广泛使用技术粗糙且缺乏约束的自研人脸识别应用 Mobile Fortify,甚至将其异化为长达 15 年的“异见者数据库”来打压抗议者。结合违规部署的 Stingray 车载伪基站等侦查设备,ICE 正利用激增的数百亿美元预算架空宪法第四修正案,将原本针对移民的监控权力迅速泛化为对全体公众的常态化监视。 ## Atmosphere & style 影片延续了 Wendover Productions 标志性的冷静、客观与抽丝剥茧的深度调查风格,旁白语调克制却暗藏锋芒。整部视频从一个科罗拉多偏远小镇的街头探头起笔,层层递进地拆解技术后门与官僚滥权,营造出一种极其真实而令人窒息的制度性惊悚感。随着视角从非法抓捕蔓延至对普通公民特权的剥夺与无死角的数字围猎,观众能真切感受到监控铁幕在日常生活中悄然降临的深沉危机感。 ## Bulletin - [00:00](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=0s) 联邦移民与海关执法局(ICE)通过 Flock Safety 车牌监控网络的跨区共享协议及商业零售探头等漏洞,成功绕过科罗拉多等蓝州的“庇护”法律获取海量行踪数据。 - [08:48](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=528s) ICE 特工广泛配备的移动人脸识别应用 Mobile Fortify 识别率低下且多次误判,但执法人员仍将其与当事人讲西班牙语等非法偏见结合,拼凑所谓“合理根据”以实施违宪逮捕。 - [11:18](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=678s) Mobile Fortify 被曝在抗议集会中无差别扫描美国公民并建档留存长达 15 年,更有法律观察员被扫描建档后随即遭撤销保持 14 年之久的 TSA PreCheck 预检资格。 - [14:00](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=840s) ICE 斥巨资部署车载 Stingray(黄貂鱼)伪基站系统,被证实存在无证调取手机识别码及在合法示威活动中实施群体性通信监控的违规乱象。 - [16:26](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=986s) 随着国会批准高达 750 亿美元的追加预算,针对边缘群体的无死角监控正在系统性侵蚀第四修正案保护,将全体公众拖入监控常态化危机。 ## Key takeaways - [00:00](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=0s) 私营监控企业 Flock Safety 的商业模式让地方警局仅拥有数据访问权而非硬件产权,其全国数据共享网络使得地方“庇护”法网被彻底穿透。 - [01:31](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=91s) 执法人员对 Flock 监控系统的查询理由缺乏实质监管,随意填入“测试”或“0”即可调取,引发多起跟踪前女友与家属的内部滥用丑闻。 - [04:07](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=247s) 即使地方警局切断对外访问权限,Lowe's 等大型商业零售卖场部署的私营 Flock 探头仍沦为 ICE 获取公众行踪的无监管后门。 - [05:53](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=353s) ICE 现场执法工具 Mobile Fortify 依托 CBP 庞大的出入境生物数据库支持,其中非美国公民的人脸照片留存期长达 75 年。 - [08:48](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=528s) Mobile Fortify 允许特工在无须取得当事人同意的情况下随时随地扫描抓拍,实战中极易因光线、抖动和回避出现严重误判。 - [08:48](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=528s) 执法特工在明知人脸算法误判的情况下,仅凭语言特征与虚假匹配拼凑“合理根据”实施抓捕,直接践踏宪法法定门槛。 - [11:18](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=678s) Mobile Fortify 生成的“遭遇日志”将面部与地理情境终身绑定,即使被采集对象是美国公民也保留长达 15 年,实质沦为打压政治异见的数据库。 - [11:18](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=678s) 抗议现场的监控数据在国土安全部(DHS)内部无缝流转,已被用于精准识别抗议者并施加撤销 TSA PreCheck 资格等定点行政报复。 - [14:00](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=840s) 单套造价 50 万美元的 Stingray 伪基站被 ICE 长期违规无证使用,甚至涉嫌在西雅图反 ICE 游行中实施无差别的群体通信劫持。 - [14:00](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=840s) 国会高达 750 亿美元的追加财政预算使 ICE 年度预算近乎翻倍,进一步加速了软硬件数字侦查分析矩阵向全美社会各角落的渗透。 - [16:26](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=986s) 执法部门缺乏动力修复技术缺陷,反而将有瑕疵的监控工具作为掩饰“先违宪抓人、后核实身份”流氓执法逻辑的制度幌子。 - [17:55](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=1075s) 原本打着针对非法移民旗号铺开的监控网,正系统性瓦解美国宪法第四修正案中关于“免受无理搜查与扣押”的基石,全面威胁全体普通公民的隐私底线。 ## Detailed walkthrough ### [00:00](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=0s) 格伦伍德斯普林斯的监控探头与 Flock Safety 的商业模式 在科罗拉多州的格伦伍德斯普林斯(Glenwood Springs),街头安装着具备 AI 识别车牌功能的监控摄像头。看似荒谬的是,美国联邦移民和海关执法局(ICE)竟能获取到这些设备的数据。格伦伍德斯普林斯在政治倾向上绝非 ICE 的支持者:在 2024 年大选中,该镇投给卡玛拉·哈里斯(Kamala Harris)的选票与对手之比达到二比一;镇上拥有庞大的移民群体,约三分之一的家庭在日常生活中使用西班牙语;而且该镇所在的科罗拉多州是一个蓝州,州法律明确禁止地方警察局与联邦移民当局共享数据或开展执法合作。然而,当地警方虽然操作并日常使用这些摄像头,但硬件产权却并不属于警局,而是归属于一家由风险投资支持的高速成长型初创企业——Flock Safety。 格伦伍德斯普林斯每年向该公司支付 3.75 万美元的订阅费以换取数据访问权限。这套配套软件功能极其强悍且立竿见影:警方不仅可以回溯特定车辆在何时、何地、是否穿过镇内街道,还能将车牌号录入追踪黑名单(Hot list),一旦该车出现在镜头中便会立刻触发警报,协助警方实施抓捕。内部日志表明,当地警方几乎每天都在利用这套系统调查从人身侵害、跟踪骚扰到性犯罪等各类案件。警局更列举了多项关键破案战果,例如逮捕一名驾驶被盗车辆的加利福尼亚州帮派分子、抓获涉嫌虐待老人的通缉重罪犯,以及截获一名从俄勒冈州诱拐儿童至该镇的恋童癖者。 ```mermaid flowchart TD accTitle: Flock Safety 监控数据流向与潜在泄露路径 accDescr: 展示格伦伍德斯普林斯监控探头数据如何在地方警局、跨州共享网络及 ICE 之间流动。 subgraph Hardware["数据采集端"] C1["Flock 车牌识别摄像头<br/>(产权归 Flock Safety)"] end subgraph Local["格伦伍德斯普林斯"] PD["格伦伍德斯普林斯警局<br/>(每年支付 3.75 万美元)"] end subgraph Network["外部共享与泄露"] Share["全国警局跨区共享协议<br/>(一键双向互通)"] ICE["联邦执法部门 (ICE / HSI)"] end C1 --> PD PD <--> Share Share --> ICE ``` ### [01:31](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=91s) 跨州数据共享网与被刺穿的“庇护”法网 格伦伍德斯普林斯的摄像头之所以极具情报价值,核心原因在于 70 号州际公路(I-70)横穿该镇,每天都有成千上万辆长途或短途车辆从监控镜头前驶过。全美各地的警方都渴望调阅这里的车牌数据,因为任何地区的本地嫌疑人逃跑时,极大概率会选择 I-70 这条大动脉以求快速逃之夭夭。为此,格伦伍德斯普林斯警局加入了遍布全美的跨区数据共享网络。只需在系统中点击几个按钮,其他外地警局就能调取格伦伍德斯普林斯的数据,反之亦然。然而,正是这种“一键互通”的机制为 ICE 彻底撕开了一道口子。 仅在 2025 年 1 月,内部使用日志就显示外部警察机构调取格伦伍德斯普林斯监控数据的次数高达 516,892 次。其中部分外地警局长期公开且主动地与 ICE 保持合作,内部日志明确记录下了国土安全调查处(HSI,即 ICE 旗下的核心调查分支机构)直接利用格伦伍德斯普林斯的数据进行案件侦查。更有甚者,该系统还显露出被用于监控政治异见人士的痕迹。在 2025 年 10 月抗议特朗普政府行动的全美“无王日”(No Kings Day)示威游行结束后的几天里,系统内出现了多起调取记录,而警局在后台填写的查询理由仅仅写着“抗议”(Protest)。尽管和平抗议活动中偶有治安违法事件,但“抗议”本身绝非合法犯罪定性,这坐实了公权力违规滥用 Flock 监控系统的风险。 - **泛滥且无约束的使用理由**:系统对警员查询的合规审查形同虚设,警员调取车牌轨迹时常填入含糊不清的理由,如“investigation”(调查)、“TBD”(待定)、“test”(测试)甚至只有一个数字“0”,导致科罗拉多方面完全无法知晓对方在查什么。 - **内部滥用与越界前科**:全美范围内执法人员私用 Flock 系统的丑闻层出不穷,堪萨斯州一名警察局长利用该系统跟踪前女友高达 228 次;南卡罗来纳州北查尔斯顿的一名警员用它监视妻子;迈阿密和威斯康星州梅纳沙的警员也分别被查出用该系统跟踪女友与前女友。绝大多数被抓包者甚至未面临刑事起诉,仅受到内部纪律处分。 ### [04:07](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=247s) 私营探头与不可控的数据暗门 当格伦伍德斯普林斯的本地居民获悉 ICE 正通过该系统渗透监控后,舆论压力迅速涌向市议会。当地警局随即关闭了科罗拉多州以外所有机构的访问权限。在理论层面上,鉴于该州禁止配合 ICE 的法律,联邦机构本该彻底失去数据来源。然而据《阿斯彭每日新闻》(Aspen Daily News)报道,随后的日志表明,科罗拉多州内部的两家警察局依然非法将数据转交给了联邦机构。迫于无奈,格伦伍德斯普林斯警局只得彻底切断该系统对外的所有访问接口。 即便本地警局采取了极端切断措施,数据安全依然千疮百孔。市政府律师明确直言,根据他们与 Flock 签署的商业合同条款,格伦伍德斯普林斯“根本无法完全控制 Flock 如何处置其系统产生的数据(Can't completely control what Flock does with data from Glenwood's system)”。这种担忧在丹佛市已成现实:丹佛在发现该市的监控数据作为某试点项目的一部分、在毫不知情且违背自身意愿的情况下被转交给联邦移民局后,果断终止了与 Flock 的合作合同。 更隐秘的盲区在于商业机构部署的监控。在格伦伍德斯普林斯的劳氏(Lowe's)家装卖场,其停车场入口处同样耸立着多台由该店自主运营的 Flock Safety 摄像头。与警局不同,私营零售商不公开任何系统日志,不对公众和选民负责,却在时刻采集进入停车场的每一辆车的车牌与轨迹。这些遍布全美的大量私营探头,已实质上沦为 ICE 在非配合性州和城市监控公民动向的无形后门。 ```mermaid flowchart TD accTitle: 商业探头与警用系统的后门通道 accDescr: 展示地方公立警局切断数据后,私营商业机构如何仍能成为监控漏洞。 subgraph Public["公立机构渠道(已受限)"] G_PD["格伦伍德斯普林斯警局"] -- "尝试切断外部链接" --> Blocked["封锁外州/跨辖区调阅"] Blocked -. "州内中介漏洞 / 合同条款".-> Leak1["数据仍存在流失隐患"] end subgraph Private["私营企业渠道(完全失控)"] Store["大型零售企业(如 Lowe's)"] PrivateCam["私营 Flock 车牌探头<br/>(全天候扫描停车场)"] PrivateCam --> Store Store --> BlackBox["无公开日志 / 商业数据协议"] end BlackBox --> ICE_Door["联邦执法部门 (ICE) 监视后门"] Leak1 --> ICE_Door ``` ### [05:53](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=353s) 从车牌雷达到面部扫描:移动抓捕工具 Mobile Fortify 民意调查显示,多数美国人对 ICE 持不信任态度。只要全美还存在愿意配合的警局,任何一个薄弱环节都会让整个车牌数据库形同虚设。但除了通过 Flock 这类平台辗转搜集行踪,ICE 还在一线大量投入专为其量身定制的定向监控工具。在近年记录 ICE 执法抓捕与对峙的现场视频中,一个极为普遍的动作是执法人员直接将手机举到被拦截者的脸前。 这并非 iPhone 的原生相机,而是由联邦移民执法部门专门研发的自研应用程序——“Mobile Fortify”。在纽约联邦法院门外,就曾有边境巡逻人员使用其变体版本对等待移民听证会的当事人进行面部扫描。一旦扫描成功,界面会立即调出该人的姓名、出生日期、国籍、当前移民身份,以及移民法官是否已签发递解出境令(Deportation order),这可直接为特工当场实施逮捕提供所谓的“合理依据”(Probable cause)。 ```mermaid flowchart LR accTitle: Mobile Fortify 识别系统数据交互逻辑 accDescr: 展示现场执法人员通过 Mobile Fortify 获取比对数据库并完成身份核验的业务流程。 subgraph Field["现场执法端"] Agent["ICE / 边境巡逻特工"] App["Mobile Fortify 应用程序"] Agent --> App end subgraph Databases["底层数据库集群"] CBP["海关及边境保卫局 (CBP)<br/>出入境人脸数据库"] ICE_Sys["ICE 案件管理系统"] FBI_NCIC["FBI 国家犯罪信息中心<br/>(NCIC)"] end App -- "现场人脸扫描图像" --> Databases Databases -- "姓名/国籍/递解令/刑事前科" --> App ``` - **海关及边境保卫局(CBP)的数据供给**:人脸识别系统的效能完全取决于其底库容量。Mobile Fortify 最核心的数据来源是隶属于美国国土安全部(DHS)的 CBP。随着 CBP 在各合法入境口岸强力推广生物识别,数据库规模呈爆炸式增长。 - **全美机场生物识别出境口**:以往美国缺乏正规的离境边检,主要依赖航空公司数据;如今越来越多的美国机场在登机口配备了生物识别闸机,通过即时人脸扫描与护照照片匹配。 - **2025 年底的全境强制推行与数据留存**:根据 DHS 在 2025 年底发布的政策,此类生物识别出境检查将扩展至所有具备国际航线的机场。美国公民在闸机采集的照片声称会在 12 小时内销毁;然而非美国公民的人脸照片将被系统存档留存长达 75 年,并全部录入至可供 Mobile Fortify 实时调阅的系统档案中。 - **多网合一的数据并联**:除 CBP 留存的合法身份者图像外,Mobile Fortify 还可以调取 ICE 自身的案件管理系统,以及联邦调查局(FBI)的国家犯罪信息中心(NCIC)数据库。 ### [08:48](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=528s) Mobile Fortify:毫无门槛的面部识别与错误抓捕 美国移民和海关执法局(ICE)的执法人员极为频繁地使用名为 Mobile Fortify 的移动应用程序。根据一份被曝光的内部文件显示,该软件之所以被滥用,根源在于其政策规定:“ICE 不向个人提供拒绝或同意收集与使用生物识别数据/照片的机会(ICE does not provide the opportunity for individuals to decline or consent to the collection and use of biometric data/photograph collection)”。这意味着特工们被允许在任何地点、任何时间、出于任何理由,对任何人直接使用该软件进行人脸识别抓拍。 然而直截了当地说,该软件的技术性能相当拙劣。2025 年 10 月,ICE 在俄勒冈州移民集中的农业重镇伍德本(Woodburn)展开了一次大规模突击搜查。随后针对 ICE 涉嫌违法行动的一起诉讼证词,揭示了 Mobile Fortify 在实战中的荒谬表现:一名执法人员扫描了一位被拘留女性的面部,随后承认“系统跳出了一个非常相似的人,但我不能确定到底是不是她”。系统最初匹配出的名字叫玛丽亚(Maria),执法人员便向该女性大喊这个名字;在未得到任何言语或肢体回应后,他再次对她进行面部扫描,结果软件竟然跳出了一个完全不同的人。 在质询笔录中,当该执法人员被问及逮捕这名女性的“合理根据”(Probable Cause——美国执法部门合法实施逮捕必须达到的宪法法定门槛)是什么时,他辩解称:“我认为她非法滞留美国,第一是因为她主要讲的语言是西班牙语,而且她和所有人交流用的都是西班牙语;第二是她与我人脸识别系统中的目标形成了疑似匹配。” 然而,美国法院早有反复判决:单凭一个人所说的语言,绝对不能构成逮捕的合理根据。换言之,该官员实施抓捕的唯一法律依据只能落在这款 Mobile Fortify 应用程序上。可就连执法人员自己也承认,系统至少产生了一次、甚至两次误判,他完全是在明知信息有误的情况下,将其作为宪法要求的逮捕理由。该诉讼目前仍在进行中,被误认女性的最终命运仍未公开,但该案折射出全美各地的 ICE 探员正在常规性地以这种漏洞百出的软件作为实施抓捕的合法借口。 ```mermaid flowchart TD accTitle: Mobile Fortify 错误识别与抓捕逻辑链 accDescr: 展示 ICE 特工如何利用存在技术缺陷的 Mobile Fortify 人脸识别软件,结合不合法的语言偏见形成虚假逮捕依据 subgraph Capture["现场人脸采集"] A["无须征求当事人同意<br/>随时随地任意抓拍"] --> B["现场条件恶劣<br/>光线差/镜头抖动/面部回避"] end subgraph Recognition["Mobile Fortify 运算"] B --> C["第一次比对<br/>匹配为 'Maria'(无应答)"] C --> D["第二次重扫<br/>跳出完全不同的人格档案"] end subgraph ProbableCause["拼凑逮捕依据"] D --> E["明知软件存在多次误判"] F["当事人与他人使用西班牙语交流"] --> G["非法拼凑'合理根据'<br/>(Probable Cause)"] E --> G G --> H["实施拘留与逮捕"] end ``` - **复杂实战环境的硬伤**:人脸识别技术在受控环境中或许准确度极高,但外勤现场往往光线不均、执法设备剧烈抖动,当事人还会转头躲避,这正是人脸识别算法最容易崩溃的场景。 - **非法程序的常态化**:由于缺乏外部监督,明知软件存在误报,特工仍将其视作完成抓捕指标的“合理工具”。 --- ### [11:18](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=678s) 跨部门监控与报复:“异见者数据库”与特权剥夺 Mobile Fortify 除了被用作非法拘禁的伪证外,其实际用途已彻底偏离了最初设定的轨道。越来越多的现场记录显示,联邦探员在各类抗议集会中高举手机,以完全符合 Mobile Fortify 操作特征的姿态扫描拍摄抗议示威者。以明尼阿波利斯(Minneapolis)的反 ICE 抗议活动为例,现场抗议者绝大多数是操英语的白人公民,特工根本不可能怀疑他们存在非法移民身份——毕竟自身面临遣返风险的人群绝不可能冒险出现在这种抗议前线。 特工拍摄抗议者的真正动机,在内部文件中露出了端倪:不同于机场通常收集的生物识别数据,Mobile Fortify 捕获的人脸图像**最长可保存 15 年,即便被拍摄对象是美国公民也同样保留**。当特工使用该应用时,后台会自动生成一条“遭遇日志(Encounter Log)”,将个人的面部图像、身份姓名与其被捕获的具体时间、地理坐标以及现场情境终身绑定。这意味着任何参与抗议公民的信息,都会在 ICE 数据库中形成长达 15 年的活动轨迹记录。 这种信息收集带来的报复性后果已有具体的司法控诉记录。在明尼苏达州的一份诉讼中,一名作为“法律观察员(Legal Observer)”跟踪并记录国土安全部(DHS)车辆执法的女性指控称:一名联邦探员突然在路中央别车逼停,使她无法前行。探员在未询问其身份的情况下直呼其名,威胁要逮捕她,并当面警告称军警部门掌握着“人脸识别技术”。仅仅三天后,该女性便收到一封官方邮件,通知她被撤销了已保持 14 年之久的运输安全管理局“预检资格(TSA PreCheck)”——这项专属于官方审核认证的“可信赖旅客”的快速安检特权凭空蒸发。 更令人担忧的是,这表明街头的人脸识别数据已经在国土安全部(DHS)内部无缝流转共享。当一名抗议者在街头被 Mobile Fortify 识别时,探员能即刻查阅其他特工此前是否对其建档,从而直接标记出此人曾出现在反 ICE 集会中。该软件实质上已演化成一个庞大的**“异见者数据库(Database of Dissent)”**,在联邦探员非法、任意拘押抗议者事件频发的背景下,沦为现场定点报复与选择性执法的决策工具。 --- ### [14:00](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=840s) 伪基站 Stingray、无证监听与全方位监控网 除了人脸识别,ICE 还装备了技术细节更为隐秘的硬件设备,例如单套造价高达 50 万美元的“黄貂鱼”(Stingray,即 IMSI 捕捉器/伪基站系统)。这类系统通常搭载于特种侦察车辆上,通过伪造基站信号来实施监控。由于现代智能手机在协议设计上会自动寻找并连接信号最强的基站,因此只要进入 Stingray 覆盖范围内的所有手机,都会被迫切换并接入该设备。 接入后,Stingray 会迅速捕获手机唯一的国际移动用户识别码(IMSI),该识别码与用户的电信运营商账户深度绑定。随后,ICE 可以通过传票、法院命令或搜查令要求运营商交出档案,将 IMSI 码与机主的真实姓名身份一一对应。尽管按照 ICE 的既定规章,鉴于 Stingray 的强大破坏力,必须在获得正式搜查令后方可针对特定目标开展“定向搜查”(例如核实某名搜捕对象究竟是在家还是在工作地点),但该机构长期公然违背自身章程,无证滥用伪基站。这一现象泛滥至极,甚至引发了国土安全部监察长办公室(Inspector General)立项发布了整份专项调查通报。 更严重的是,已有间接证据表明该机构涉足无差别的群体性大规模监控。2025 年 7 月 4 日,在华盛顿州西雅图举行的一场反 ICE 抗议游行中,现场蜂窝通信网络信号分析发现了与 IMSI 捕捉器运作完全吻合的异常波动。这表明 ICE 极有可能在集会现场部署了伪基站,无差别吸取现场每一位示威者的手机识别码,从而建立起完整的集会人员数字点名册。 ```mermaid flowchart LR accTitle: Stingray 伪基站手机监控流程 accDescr: 展示黄貂鱼伪基站如何劫持手机信号并获取用户真实身份,乃至用于无差别抗议现场监控 subgraph Spoof["信号劫持阶段"] A["车载 Stingray 伪基站<br/>(单套造价约 $500,000)"] -- "发射强欺骗信号" --> B["区域内所有智能手机"] B -- "被迫接入" --> C["强制抓取 IMSI 识别码"] end subgraph Link["身份落地与违规应用"] C --> D["比对电信运营商账户档案"] D --> E["精确锁定公民真实身份"] E --> F["合规模式:持搜查令定向追踪"] E --> G["违规乱象:无证使用 & 游行现场群体建档"] end ``` - **庞大的软硬件矩阵**:车牌识别系统(ALPR)、Mobile Fortify 与 Stingray 仅是冰山一角。ICE 坐拥庞大的数字侦查分析软件套件,不断将案情挖掘自动化,从而腾出大批外勤警力在街头实施大规模围捕。 - **激增的财政拨款**:国会近期通过了高达 750 亿美元的追加预算,使得 ICE 未来几年的年度预算近乎翻了三倍,其监控系统的广度与下沉深度正以惊人的速度扩张。 --- ### [16:26](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=986s) 宪法危机的延伸:被系统性侵蚀的第四修正案 虽然上述许多技术漏洞百出、错误率极高,但 ICE 几乎没有任何技术升级和排错的动力。因为就其真实目标而言,这些工具非常“好用”——由于在日常环境中很难为每名嫌疑人搜集到扎实合法的法定逮捕依据(Probable Cause),这些有缺陷的监控工具便顺理成章地被拿来充当幌子,用来掩盖无凭无据的非法拘捕。执法人员的标准流程变成了:先践踏宪法权利将人扣押,再慢慢调取后台核实身份。在当前缺乏实质权力约束的环境下,这种违宪套路屡试不爽。 许多美国公众长期以来对移民的人权与宪法权利漠不关心,甚至默许这种违法执法的提速。然而,这种监控逻辑的扩散早已演变成对全体公民隐私的根本威胁。ICE 正在推动一场席卷全美的常态化监控扩张(Mass Surveillance Creep),其影响贯穿多届政府,并在当下一路狂飙。 历史上,美国社会对本土政府监控保持着极高警惕,民调显示美国公众对隐私受侵害的担忧远超欧洲平均水平,宁愿牺牲一定程度的安全也要捍卫个人隐私权。美国宪法第四修正案中“免受无理搜查与扣押”的核心保护伞,正在被 ICE 的技术铁幕彻底击碎——通过对整个物理空间和数字信号实施无死角的无差别扫描,它实质上是在不间断地“搜查我们当中的每一个人”。 ### [17:55](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=1075s) 监控常态化与监控国家的悄然降临 ICE 等机构往往在没有搜查令(warrant)的情况下四处刺探以期捕获有用线索。表面上看,许多人可能觉得这种手段情有可原,甚至是合情合理的;但大量且压倒性的事实证明,该机构在行事方便时会随意忽视或修改自身政策,极力对公众隐瞒具体侦察手段,根本无法让人信任他们能够以所谓“合规、正当的方式”进行监控。 这种随意且毫无节制的执法手段,有时确实提高了驱逐移民的效率,从而为某些人提供了辩解的说辞。然而,随之而来的严重代价是将一种极其危险的现状合理化与常态化——在这个国家里,当权者拥有了随时随地知晓“你在哪里、你认识谁、你长什么样以及你正在做什么”的能力。这种监控权力已经在被滥用;眼下这种针对移民和抗议者的侵害行为,未来完全可能轻而易举地转向任何一个普通人。美国正在蜕变为一个无孔不入的“监控国家(surveillance state)”,而太多人之所以尚未感到担忧,仅仅是因为眼下的受害者主要被官方贴上了“敌人”的标签。 ```mermaid flowchart TD accTitle: 监控权力泛化机制示意图 accDescr: 展示监控手段如何从针对特定群体逐步外溢至全体公众的风险链路 subgraph S1["现有实践"] A["绕过搜查令与隐蔽手段"] --> B["提升驱逐移民与打压抗议效率"] end subgraph S2["系统性外溢"] B --> C["权力滥用常态化<br/>(知晓位置、人际网络与行动)"] C --> D["监控对象泛化:最终波及全体普通公民"] end ``` - **滥用监控的扩散效应**:权力一旦被默许用于对付特定被边缘化的群体,其技术框架与执法惯性便会永久成型,最终任何人都可能成为监视对象。 - **缺乏监督与问责**:相关机构在政策执行上频繁出现“暗箱操作”,对外掩盖手段,对内选择性执行规则,彻底动摇了公众对程序正义的信任。 --- ### [18:49](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=1129s) 数字隐私防线与 NordVPN 赞助 现实中许多形式的数字监控让人防不胜防,但并非所有数据泄露都无可挽回。举例来说,普通用户使用的互联网服务提供商(ISP)通常会记录相当数量的网页浏览历史,在美国,只要政府部门出具正规的传票或搜查令,这些浏览数据就可能被直接移交。虽然许多人平时对此不以为意,但这正是数据隐患的典型缩影,也体现出使用虚拟专用网络(VPN)的价值所在。 开启 NordVPN 后,系统会在用户与远程服务器之间构建一条高度加密的隧道。与许多会留存痕迹的 ISP 不同,NordVPN 严格执行无日志政策,即使面临政府机构的强制传唤,他们也拿不出任何可以移交的用户数据;这一安全声明已由普华永道(PwC)与德勤(Deloitte)等独立第三方会计师事务所完成专项审计确认。 除了防范潜在的政府窥探,这类工具在日常网络生活中同样具备多重实用功能: - **绕过网络防火墙与内容限制**:在遇到特定网站被屏蔽的地区或旅行时,用户可将流量虚拟重定向至分布在众多国家的节点服务器,自由访问开放网络。 - **破解流媒体与网站的地理封锁(Geo-blocking)**:当视频流媒体服务因地域限制无法播放特定内容时,只需轻点几次切换节点即可解决。 - **日常安全防护(Threat Protection Pro)**:内置的威胁防护系统能够主动拦截高危广告、弹窗及恶意网站,显著降低设备感染恶意软件的风险。 - **Wendover 专属优惠权益**:通过官方合作链接 `nordvpn.com/wendover` 注册可享受额外赠送 4 个月的专属优惠,并附带 30 天无理由退款保证以实现零风险试用。 ## Notable quotes > [04:07](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=247s) 根本无法完全控制 Flock 如何处置其系统产生的数据 — 格伦伍德斯普林斯市政府律师 > [08:48](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=528s) ICE 不向个人提供拒绝或同意收集与使用生物识别数据/照片的机会 > [08:48](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=528s) 系统跳出了一个非常相似的人,但我不能确定到底是不是她 — ICE 执法人员 > [08:48](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=528s) 我认为她非法滞留美国,第一是因为她主要讲的语言是西班牙语,而且她和所有人交流用的都是西班牙语;第二是她与我人脸识别系统中的目标形成了疑似匹配。 — ICE 执法人员 > [11:18](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=678s) 最长可保存 15 年,即便被拍摄对象是美国公民也同样保留 > [11:18](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=678s) 异见者数据库 > [16:26](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=986s) 免受无理搜查与扣押 — 美国宪法第四修正案 > [16:26](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=986s) 搜查我们当中的每一个人 > [17:55](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=1075s) 你在哪里、你认识谁、你长什么样以及你正在做什么 > [17:55](https://www.youtube.com/watch?v=aRUhgrqnpGY&t=1075s) 监控国家 ## Mentioned Wendover Productions, Sam Denby, 美国联邦移民和海关执法局 (ICE), 国土安全调查处 (HSI), 美国国土安全部 (DHS), 美国海关及边境保卫局 (CBP), 联邦调查局 (FBI), 国家犯罪信息中心 (NCIC), 运输安全管理局 (TSA), TSA PreCheck, Flock Safety, Mobile Fortify, 黄貂鱼 (Stingray), 格伦伍德斯普林斯 (Glenwood Springs), 卡玛拉·哈里斯 (Kamala Harris), 唐纳德·特朗普 (Donald Trump), 劳氏 (Lowe's), 丹佛 (Denver), 伍德本 (Woodburn), 明尼阿波利斯 (Minneapolis), 西雅图 (Seattle), 《阿斯彭每日新闻》 (Aspen Daily News), NordVPN, 普华永道 (PwC), 德勤 (Deloitte) ## Video description Get Exclusive NordVPN deal + 4 months extra here → https://nordvpn.com/wendover It’s risk-free with Nord’s 30-day money-back guarantee! Sponsorship Enquiries: wendover@standard.tv Other emails: sam@wendover.productions Reddit: http://Reddit.com/r/WendoverProductions Writing by Sam Denby Editing by Alexander Williard Animation by Sara Stoltman, Gabriel Ferreras, Austin Glass, and Kate Ermolenkko Sound by Manni Simon and Dony Bullen Thumbnail by Simon Buckmaster References [1] https://www.cogs.us/DocumentCenter/View/10579/Flock-Safety_successful-uses [2] https://local12.com/news/nation-world/police-chief-gets-caught-using-license-plate-cameras-to-track-his-ex-girlfriend-228-times-arrests-charges-probation-flock-safety-follow-stalk-new-boyfriend-broke-up-out-of-town-misuse; https://www.kcrg.com/video/2024/06/24/video-north-charleston-police-officer-demoted-using-cameras-track-wife/; https://www.miamiherald.com/news/state/florida/article299881644.html; https://fox11online.com/news/crime/menasha-officer-pleads-not-guilty-to-misconduct-charge-over-ex-girlfriend-tracking-cristian-morales-license-plate-recognition-flock-restraining-order-domestic-abuse-case?utm [3] https://prri.org/spotlight/most-americans-distrust-u-s-immigration-and-customs-enforcement-ice-and-believe-federal-funding-increases-have-gone-too-far/?utm [4] https://www.documentcloud.org/documents/26209262-mobile-fortify-pta/?ref=404media.co&q=opportunity&mode=document#document/p4 [Source](https://www.youtube.com/watch?v=aRUhgrqnpGY)

Source quotes

[4:07] 格伦伍德斯普林斯市政府律师: 根本无法完全控制 Flock 如何处置其系统产生的数据 [8:48] ICE 不向个人提供拒绝或同意收集与使用生物识别数据/照片的机会 [8:48] ICE 执法人员: 系统跳出了一个非常相似的人,但我不能确定到底是不是她 [8:48] ICE 执法人员: 我认为她非法滞留美国,第一是因为她主要讲的语言是西班牙语,而且她和所有人交流用的都是西班牙语;第二是她与我人脸识别系统中的目标形成了疑似匹配。 [11:18] 最长可保存 15 年,即便被拍摄对象是美国公民也同样保留 [11:18] 异见者数据库 [16:26] 美国宪法第四修正案: 免受无理搜查与扣押 [16:26] 搜查我们当中的每一个人 [17:55] 你在哪里、你认识谁、你长什么样以及你正在做什么 [17:55] 监控国家

Source quotes

[6:29] Internal ICE Document: ICE does not provide the opportunity for individuals to decline or consent to the collection and use of biometric data/photograph collection. [9:18] ICE Officer: it came up to a very like a similar person. I wasn't sure if it was her or not. [9:18] ICE Officer: became a possible match on my facial recognition system. [17:14] America is becoming a surveillance state, and too many aren't concerned because the victims so far are primarily those they've been told are the enemy.

Original transcript

[0:00] This is an AI-enabled license plate [0:03] reading surveillance camera, and in a [0:05] convoluted way, the US's federal [0:07] Immigration and Customs Enforcement [0:08] Agency, more commonly known as ICE, has [0:11] been able to access its data. This [0:13] despite that the town whose police [0:14] department operates this camera, [0:16] Glenwood Springs, is about as far from [0:18] supportive of ICE as any. Glenwood voted [0:21] two to one for Kamala Harris in the 2024 [0:23] election and is home to a large [0:24] immigrant population with about a third [0:27] of residents speaking Spanish within [0:28] their household. It's also in Colorado, [0:30] whose majority blue voter base supported [0:32] the passage of a statewide law [0:34] explicitly prohibiting local police [0:35] departments from sharing data or [0:37] cooperating with federal immigration [0:38] authorities. But while this camera is [0:41] operated by the local police department [0:42] and primarily used by the local police [0:44] department, it's not actually owned by [0:46] them. It's owned by a fast-growing [0:48] venture capital-backed startup called [0:50] Flock Safety. The town then pays an [0:52] annual fee of 37 and 1/2 thousand [0:54] dollars to be able to access its data, [0:56] [music] and that data is apparently [0:58] useful. Through the included software [1:00] suite, they're able to go back in time [1:02] and check where, when, and if a [1:03] particular vehicle passed through their [1:05] streets or add a license plate to a hot [1:07] list to get an alert when it appears, [1:09] allowing for the occupants' arrest. [1:11] Internal logs show the department uses [1:13] the cameras just about daily to [1:14] investigate anything from assault to [1:16] stalking to sex offenses. And Glenwood [1:19] PD purports these searches have been [1:20] instrumental in a number of successful [1:22] cases like arresting a gang member from [1:24] California in a stolen vehicle, a wanted [1:26] felon suspected of elder abuse, or a [1:28] pedophile who lured a child to the town [1:30] from Oregon. [1:31] What makes Glenwood's cameras more [1:33] useful than most is the fact that [1:34] Interstate 70 passes through the town, [1:36] meaning [music] thousands of cars on [1:38] journeys both long and short drive [1:39] through their view each day. [1:41] >> [music] [1:41] >> That's why police departments from [1:42] around the country want access to their [1:44] data. There's every chance a local [1:46] suspect from just about anywhere fled [1:48] via Interstate 70 in an attempt to get [1:50] far away fast. So Glenwood opted in to [1:53] data sharing agreements with police [1:55] departments around the country. [1:56] Basically, for the press of a couple [1:58] buttons, other police departments could [1:59] now search Glenwood's data, and Glenwood [2:01] could do the same with theirs. But, [2:03] that's how ICE got in. [2:06] In January 2025 alone, internal usage [2:08] logs indicate that outside police [2:10] departments accessed Glenwood's data [2:12] 516,892 [2:15] times. A subset of these outside police [2:17] departments do actively and openly [2:19] cooperate with ICE, which explains logs [2:22] like these. Explicit acknowledgement [2:24] that HSI, ICE's investigative arm, is [2:27] conducting a search using Glenwood's [2:29] data. [2:30] There's even indication that Glenwood's [2:31] data might have been used to track [2:33] political dissent. In their October [2:35] logs, there are a number of searches in [2:37] the days following the No Kings Day [2:38] protest, a nationwide demonstration [2:40] against the actions of the Trump [2:41] administration, logged with the [2:43] reasoning protest. Legitimate crimes do [2:45] happen even at overwhelmingly peaceful [2:47] protests, but the reasoning logged by [2:49] police departments was not one, meaning [2:51] this at the least demonstrates their [2:53] potential to use Flock data in [2:55] illegitimate ways. That's especially [2:58] true since there's really no standard as [3:00] to how much police departments, or even [3:02] individual officers, must share to [3:04] justify why they're using Flock data. [3:06] There are thousands upon thousands of [3:08] instances each month where Glenwood's [3:10] Police Department has no idea what other [3:12] agencies are doing with their data since [3:14] those agencies log the reason for their [3:16] search as something ambiguous like [3:18] investigation, TBD, test, or the number [3:21] zero. These agencies could be doing [3:23] whatever with Glenwood's data, and the [3:25] Colorado would have no idea. [3:29] And whatever certainly has happened. [3:31] There's plenty of precedent of nefarious [3:33] use by individual rogue actors within [3:35] law enforcement agencies. A police chief [3:38] in Kansas was caught using Flock data to [3:40] stalk his ex-girlfriend 228 times. A [3:43] North Charleston, South Carolina officer [3:45] did the same with his wife. One in Miami [3:47] did so with his girlfriend. [music] [3:48] One in Menasha, Wisconsin did with his [3:50] ex. The cases go on and on, and it's [3:52] hardly surprising given so many of those [3:54] caught don't even face legal action and [3:56] are simply disciplined internally. And [3:59] of course, [music] these are only the [4:00] cases that are caught, and given the [4:02] spotty tracking of Flock usage, there's [4:04] certainly ample ability to fly under the [4:06] radar. [4:07] Now, eventually local residents in [4:09] Glenwood Springs got wind of ICE's [4:10] access to their local data and put [4:12] pressure on the city council. Their [4:14] police departments then closed off [4:15] access to the system for agencies [4:17] outside Colorado, meaning in theory, ICE [4:19] should no longer have had access given [4:21] the state law prohibiting cooperation [4:23] with ICE, but still then, according to [4:25] reporting in the Aspen Daily News, logs [4:27] show that two police departments within [4:29] Colorado illegally shared data with the [4:31] federal agency. In response, Glenwood [4:34] shut off outside access to the system [4:36] entirely, but still then, the city's [4:38] attorney stated that the terms of their [4:40] contract with Flock meant they quote, [4:42] "Can't completely control what Flock [4:43] does with data from Glenwood's system." [4:46] And there's precedent supporting the [4:47] idea that the data's still not secure. [4:50] Three hours down the road, Denver [4:51] terminated their contract with Flock [4:53] after it came out that the city's data [4:54] was shared with federal immigration [4:56] authorities against their wishes or [4:58] knowledge as part of a pilot program. [5:00] And in Glenwood, even if the police [5:02] department's cameras are successfully [5:04] walled off from the outside world, that [5:06] doesn't mean that ICE can't get access [5:08] to Flock safety data from the town. [5:11] This is the local Lowe's Home [5:12] Improvement store, and here, at the [5:14] entrance to its parking lot, are even [5:17] more Flock safety cameras. These cameras [5:20] are operated and used by the store [5:22] itself. Unlike the local police [5:23] department, Lowe's doesn't publish its [5:25] logs, it doesn't say who's got access to [5:28] its data, it's collecting a database of [5:30] every single car that enters its lot, [5:32] and the public has absolutely no idea [5:34] what it's doing with that data. It's [5:36] just one of thousands of Flock safety [5:39] cameras around the country not operated [5:41] by public agencies that must answer to [5:43] their constituents, but by private [5:45] companies, meaning they could act as a [5:47] backdoor for ICE to surveil the [5:48] movements of Americans even in [5:50] non-cooperative states [music] and [5:51] municipalities. [5:53] Now, in recent polling, a majority of [5:55] Americans have indicated distrust in [5:57] ICE. So, even if Flock Safety is not a [6:00] system the agency [music] created, their [6:01] access to local surveillance data is a [6:03] growing cause of concern. Plenty, [6:05] perhaps most police departments have not [6:07] received the pressure that Glenwood's [6:09] has and are still sharing data with [6:11] essentially any agency that asks. Any [6:13] database is only as secure as its [6:15] weakest link, and there are some very [6:17] weak links with this one, [6:19] >> [music] [6:19] >> meaning ICE now has fairly open access [6:21] to an effectively nationwide network of [6:23] license plate surveillance cameras. All [6:25] they need to do is find one friendly [6:27] local police department. But, of course, [6:30] many of ICE's surveillance tools don't [6:32] require roundabout access. Many were [6:34] designed specifically for the agency's [6:36] use. In any of the endless videos [6:38] documenting tense, sometimes violent ICE [6:40] arrests or confrontations, there's an [6:42] increasingly common sight: agents [6:45] holding a phone up to a person's face. [6:47] This has a very specific purpose, and [6:50] this photo reveals it. This isn't the [6:53] iPhone's camera app. It's a proprietary [6:55] one developed specifically for use by [6:57] federal immigration agents called Mobile [6:59] Fortify. Here it can be seen in action. [7:02] This agent is attempting to use a Border [7:04] Patrol variant of it to identify this [7:06] individual waiting for an immigration [7:07] hearing at a federal courthouse [music] [7:08] in New York. If the agent scan is [7:10] successful, he'll see the individual's [7:12] name, date of birth, nationality, [7:14] immigration status, and whether a judge [7:16] has already ruled that they should be [7:17] deported, potentially giving [music] the [7:19] agent probable cause for an arrest. Of [7:21] course, like any facial recognition [7:23] software, this app is only as good as [7:25] its imagery database. For that, ICE [7:28] siphons data primarily from Customs and [7:30] Border Protection, another agency under [7:32] the DHS umbrella that deals with [7:34] processing people at legal ports [music] [7:35] of entry. Over recent years, CBP has [7:38] dramatically ramped up their use of [7:40] biometrics, giving them a massive, yet [7:43] still growing database of identity data [7:45] on Americans [music] [7:46] and non-Americans alike. You see, [7:48] previously, unlike many countries, the [7:50] US had no formal [music] exit [7:52] immigration controls. Rather, airlines [7:55] shared data with authorities, which gave [7:56] them a mostly accurate picture of who [7:58] left the US and when. Today, though, an [8:01] increasing number of US airports are [8:03] equipped with a biometric gates that [8:04] scan people's faces and match them [8:06] against their passport photos when [8:08] boarding a plane. This is both more [8:10] accurate and new imagery of an [8:12] individual. Towards the end of 2025, DHS [8:15] announced that these exit immigration [8:17] checks would become universal, meaning [8:19] biometric gates are in the process of [8:20] being installed at any airport with [8:22] international flights, dramatically [8:24] increasing the amount of data the agency [8:25] will collect. While photos from these [8:27] gates of US citizens are said to be [8:29] deleted within 12 hours, non-citizens' [8:32] photos are kept for up to 75 years and [8:35] put into a file that can be accessed by [8:37] ICE's Mobile Fortify app. This applies [8:39] to much of the imagery that identifies [8:41] those with legal status in the [music] [8:42] US, while Mobile Fortify also has access [8:45] to ICE's case management system and the [8:47] FBI's [music] National Crime Information [8:48] Center to provide imagery of individuals [8:50] more likely to generate probable cause [8:52] for arrest. [8:54] ICE agents use Mobile Fortify a lot, and [8:57] this internal document indicates why. [8:59] Quote, ICE does not provide the [9:01] opportunity for individuals to decline [9:03] or consent to the collection and use of [9:05] biometric data/photograph collection. [9:08] Basically, agents are allowed to use the [9:10] app on anyone, anywhere, [music] at any [9:12] time, for any reason. This is a source [9:15] of problems, since, to put it bluntly, [9:18] the app isn't very good. In October [9:21] 2025, ICE agents conducted a major [9:23] operation in the immigrant-heavy [9:25] agricultural town of Woodburn, Oregon. [9:27] >> [music] [9:28] >> A deposition from a lawsuit related to [9:29] alleged unlawful actions by ICE during [9:31] this raid gives a glimpse into how [9:33] Mobile Fortify actually functions in the [9:35] field. An officer scanned the face of [9:37] one of the detained individuals and, [9:39] quote, "It came up to a very like a [9:41] similar person. I wasn't sure if it was [9:43] her or not." The match was for a person [9:45] named Maria, so he called out Maria and [9:48] given no response, verbal or otherwise, [9:50] he scanned her face again. Then the app [9:53] matched with an entirely different [9:55] person. In the deposition, the officer [9:58] was then asked what his probable cause [10:00] was for arresting this woman, the [10:02] constitutionally mandated threshold [10:04] American law enforcement must pass to [10:06] lawfully arrest an individual. The [10:08] officer responds, quote, "I believe she [10:10] was in the country illegally because [10:11] one, the primary language she used that [10:13] I heard her use was Spanish, and then to [10:15] interact with everyone else was Spanish. [10:17] She became a possible match on my facial [10:19] recognition system." [10:21] Courts have repeatedly ruled that the [10:22] language an individual speaks alone [10:24] cannot be used as probable cause, [10:27] meaning the court legal probable cause [10:29] the officer might have been using to [10:31] justify the arrest was the Mobile [10:33] Fortify app. And of course, the officer [10:36] himself admitted that the app [10:38] misidentified the woman at least once, [10:40] potentially twice, meaning he was solely [10:42] relying on knowingly faulty information [10:45] as the basis of his probable cause. This [10:47] case is still ongoing and it's not [10:49] publicly known what happened to the [10:50] misidentified woman, but what is known [10:53] is that across the country, ICE agents [10:55] are using this faulty app as the basis [10:58] of their probable cause. While facial [11:00] recognition might be highly accurate in [11:02] a controlled environment, the settings [11:04] in which agents are using Mobile Fortify [11:06] are far from that. Lighting is [11:08] inconsistent, the camera is unstable, [11:10] individuals are moving their head to [11:12] evade scans. It's the exact situation [11:15] where facial recognition struggles. [11:18] But beyond the problem of the app [11:19] functioning as a false premise to [11:21] justify unlawful detentions, the app is [11:24] also being used in a manner entirely [11:25] different than intended. Increasingly, [11:28] federal agents have been seen holding [11:29] their phones up, capturing imagery of [11:31] protesters in a manner consistent with [11:33] the use of Mobile Fortify. In settings [11:35] like Minneapolis, anti-ICE protesters [11:37] were overwhelmingly white, [11:39] English-speaking individuals, meaning [11:40] the agents would not likely suspect them [11:42] of irregular immigration status. [11:44] Clearly, people vulnerable to legal [11:46] deportation by ICE are unlikely to go [11:48] out into such a setting. So, why would [11:50] agents be using Mobile Fortify against [11:52] these protesters? Well, while there's no [11:54] way to know for sure, this internal [11:56] document hints at a theory. [11:58] Unlike the airport biometric data, this [12:00] says that imagery from Mobile Fortify is [12:02] retained for up to 15 years, even if the [12:05] pictured individual is a US citizen. [12:08] What we do know with certainty is that [12:10] when an agent uses Mobile Fortify, a [12:12] so-called encounter log is created by [12:14] the app where an individual's photo and [12:16] name is tied to the time, location, and [12:18] circumstance of its capture. So, that's [12:20] to say, if an individual's image is [12:22] captured at a protest, ICE now has an [12:24] electronic record for the next 15 years [12:27] that that individual was there. And [12:29] there's anecdotal evidence of what the [12:30] federal government does with this [12:31] information. This court filing related [12:34] to the agency's actions in Minnesota [12:35] includes an allegation by a woman acting [12:37] as a so-called legal observer, following [12:40] DHS vehicles and documenting their [12:41] actions. She alleges that the agent, [12:43] quote, "stopped in the road, making it [12:45] impossible for her to proceed." The [12:47] officer then addressed the woman by [12:49] name, despite not having volunteered her [12:51] name, threatened her with arrest if she [12:53] continued, and warned her that they had [12:55] facial recognition technology. [12:57] Three days later, she got an email [12:58] saying that the TSA had revoked her [13:00] precheck status, the widespread program [13:03] that allows for expedited airport [13:04] screening for so-called trusted [13:06] travelers. Having been a member of the [13:07] program for 14 years, one can reasonably [13:10] assume that the two events were [13:11] connected. While this appears to be an [13:13] isolated incident for now, what it [13:15] demonstrates is that on the street [13:17] facial recognition data is flowing [13:19] elsewhere within DHS. And even if [13:21] there's no widespread systematic effort [13:23] of retribution against anti-ICE [13:25] protesters yet, given what we know about [13:27] how MobileFortify works, there's a [13:29] reason to believe that this is already [13:31] happening on a small scale out in the [13:33] field. It's believed that when an [13:35] individual is identified using [13:37] MobileFortify, agents can see if another [13:39] already used the app on that individual. [13:42] That means they can see the indicators [13:44] that someone was previously at an [13:45] anti-ICE protest, meaning it acts as a [13:48] database of dissent. Unlawful and [13:50] arbitrary detention of protesters by [13:52] federal agents is well reported and [13:54] widespread, and the app could act as a [13:56] decision-making tool of who agents want [13:59] to target. Now, with each additional [14:01] surveillance tool ICE is known to [14:02] possess, the details get fuzzier. For [14:05] example, we know they have these, [14:08] stingrays. These are typically included [14:10] as part of a half-million-dollar [14:11] vehicle-based cell phone surveillance [14:13] system, and they work by effectively [14:15] spoofing a real cell signal. Cell phones [14:17] are designed to constantly switch to the [14:19] strongest signal, typically from the [14:20] closest tower, meaning any cell phone [14:23] near a stingray device will connect. [14:25] Next, the device is able to capture this [14:27] connected cell phone's IMSI number, a [14:29] unique numerical combination that [14:30] corresponds to the user's carrier [14:32] account. Through a subpoena, court [14:34] order, or warrant, ICE is then able to [14:36] gain access to carrier records and draw [14:38] a link between an IMSI number and a [14:40] user's identity. Considering how [14:42] powerful stingrays are as a surveillance [14:44] tool, ICE's policies only allow targeted [14:47] use with a warrant. Essentially, [music] [14:49] it must have a directed purpose towards [14:51] finding a specific device or individual. [14:54] Of course, this can be legitimately [14:56] useful for determining whether a [14:58] targeted individual is at home or at [15:00] work, for example. There is, however, a [15:02] long, long history of ICE ignoring its [15:06] own policies and using stingrays without [15:08] a warrant. It's so pervasive, in fact, [15:11] that there's a whole Inspector General [15:13] report about the issue. But, there's [15:15] also some circumstantial evidence of the [15:17] agency even more flagrantly breaking [15:19] policy and using stingrays [music] in [15:21] non-targeted mass surveillance [15:23] operations, capturing and collecting the [15:25] IMSI numbers of anyone [music] in a [15:28] target area. Analysis of cellular [15:30] signals at a July 4th, 2025 anti-ICE [15:33] protest in Seattle, Washington, found [15:35] anomalies consistent with IMSI catcher [15:37] use, meaning the agency might have been [15:39] using devices to gather a log of all the [15:41] devices and therefore individuals [15:43] present at that demonstration. License [15:46] plate readers, mobile fortify, and [15:48] stingrays [music] really are only the [15:49] tip of the iceberg. ICE also has access [15:52] to a massive software suite of digital [15:54] surveillance tools that are increasingly [15:55] automating their investigative work, [15:57] allowing for more personnel to stay out [15:59] in the streets rounding up more [16:01] individuals for deportation. There's [16:03] also all the tools we don't know about. [16:05] ICE rarely volunteers information about [16:07] their surveillance systems, so what we [16:09] know comes from leaks, court documents, [16:11] mandatory disclosures, and observation. [16:14] With $75 billion in additional funding [16:16] recently approved by Congress, nearly [16:18] tripling ICE's annual budget over the [16:20] next few years, the depth and breadth of [16:22] the agency's systems are sure to expand [16:24] increasingly fast. This is, of course, a [16:27] threat to immigrant rights. Many of [16:29] these tools just aren't that effective. [16:31] They have massive error rates, but ICE [16:33] has little motivation to improve these [16:35] ineffective tools since for their [16:37] purposes, they're effective. They can be [16:39] used as a pretext for unlawful arrests [16:42] of those lacking legal status without [16:44] legitimate probable cause. It is [16:46] difficult for the agency to gain [16:48] legitimate probable cause on [16:49] individuals, so it's simply easier for [16:52] them to ignore people's constitutional [16:54] rights, detain them, then investigate [16:56] their residency status. With no one in [16:58] power in the business of challenging [17:00] ICE's lawfulness these days, this just [17:02] works. And many Americans have chosen to [17:04] not concern themselves with the [17:06] constitutional and human rights of [17:07] immigrants so long as doing so hastens [17:09] the administration's progress towards [17:11] their goal of a more homogeneous [17:12] America. For those people, however, the [17:15] means towards ICE's ends should still be [17:18] a source of concern. The agency has [17:20] played a meaningful role in a massive [17:22] creep in mass surveillance of the [17:24] American people. One that has happened [17:26] government wide across administrations, [17:28] but one that ICE has helped hasten. [17:31] Traditionally, the US public has been [17:33] uniquely wary of domestic government [17:35] surveillance. Polling indicates that [17:37] Americans are far more concerned by this [17:39] than the average European, for example, [17:41] and are generally content with a certain [17:43] compromise in security in exchange for [17:45] increased privacy. One of the most [17:47] important parts of the Constitution is [17:48] the protections against unreasonable [17:50] search and seizure. ICE is challenging [17:53] this by effectively searching everyone [17:55] without warrants in hopes of finding [17:57] useful info. Many might consider the [18:00] surface level of their use worthwhile, [18:02] but there's ample overwhelming evidence [18:04] that the agency ignores or changes [18:06] policy when convenient, hides their [18:07] methods from the public, and generally [18:09] can't be trusted to do surveillance the [18:11] so-called right way. [18:13] >> [music] [18:13] >> At times these fast and loose methods [18:15] make the agency more efficient at [18:16] removing immigrants, meaning some might [18:18] justify them, but at other times it [18:20] normalizes an America where people in [18:22] power have the ability to know where you [18:24] are, who you know, what you look like, [18:27] and what you're doing at any given time. [18:30] This power is already being abused, but [18:32] the same abuse that's currently being [18:33] targeted at immigrants and protesters [18:36] can later be directed towards simply [18:38] anyone. America is becoming a [18:40] surveillance state, and too many aren't [18:42] concerned because the victims so far are [18:44] primarily those they've been told are [18:47] the enemy. [18:49] Many of these forms of digital [18:50] surveillance are effectively [18:52] unavoidable, but not all. For example, [18:55] your internet service provider likely [18:57] keeps some amount of data on your [18:58] browsing history, and in the US, [music] [19:00] this could be handed over to the [19:02] government if they have a proper [19:03] subpoena or warrant. While this is not [19:05] too much of a concern for most, it's [19:07] just one of almost endless examples of [19:09] where our sponsor NordVPN would be [19:12] useful. Turning their VPN on creates an [19:14] encrypted tunnel between yourself and [19:16] their VPN server, [music] and unlike [19:18] many internet service providers, Nord [19:20] VPN doesn't keep any usage data that [19:22] they could hand over to a government [19:24] even if they were compelled to. They've [19:26] even been independently reviewed by PwC [19:28] and Deloitte to assure that this is [19:29] really true. But beyond evading [19:31] government surveillance, Nord VPN has [19:33] plenty other practical applications, [19:35] too. [music] For example, it can also [19:37] get around government firewalls. If [19:39] you're traveling or living in a place [19:40] where certain sites are blocked, Nord [19:42] VPN allows you to virtually redirect [19:44] your traffic through one of their [19:45] servers in any number of countries, [19:47] allowing you to browse a more open [19:49] version of the internet. The same works [19:51] for geo-blocking on websites. If your [19:53] streaming service isn't allowing you to [19:54] watch something based on where you're [19:56] browsing from, Nord VPN can fix that [19:58] with just a few clicks. It's also great [20:00] for day-to-day internet security. Their [20:02] threat protection pro system [20:04] automatically blocks dangerous ads, [20:05] pop-ups, or websites, reducing the risk [20:08] of getting malware. The uses go on, but [20:10] the point is that, in my opinion, [20:11] [music] a VPN is a crucial tool for any [20:14] internet user today. I find myself using [20:16] Nord VPN for something on pretty much a [20:18] daily basis. If you'd like to sign up, [20:20] they're offering an exclusive deal for [20:21] Wendover viewers, including 4 months [20:23] extra when you go to [20:24] nordvpn.com/wendover [20:26] or click the link in the description. [20:28] And it's worth giving them a shot since [20:30] it's risk-free with their 30-day [20:32] money-back guarantee.