A recently published investigation has revealed the existence of secretive units within U.S. Border Patrol that reportedly analyze Americans’ financial activity and other data, then share that intelligence with local law enforcement who conduct traffic stops. The units are called Predictive Intelligence Targeting Teams, or PITT.
The investigation, published by technology news outlet 404 Media on September 8, 2026, identified PITT teams operating in at least two Border Patrol sectors—Spokane, Washington, near the Canadian border, and Laredo, Texas, near the Mexican border. According to the reporting, these teams review what a Department of Homeland Security document describes as “law enforcement-sensitive databases” to develop intelligence on individuals who are not suspected of any specific crime.
The program has drawn scrutiny from civil-liberties advocates who question whether the practice of using financial data to generate traffic stops violates constitutional protections against unreasonable searches and seizures. U.S. Customs and Border Protection, which oversees Border Patrol, has defended its intelligence practices as lawful while declining to discuss specific operational details.
This explainer examines what PITT teams are, what the available reporting establishes, what CBP says about the program, and what important questions remain unanswered.
Predictive Intelligence Targeting Teams are specialized units within U.S. Border Patrol’s Targeting and Intelligence Divisions. According to documents reviewed in the 404 Media investigation, PITT personnel are assigned to review law-enforcement databases and develop intelligence that can be passed to local police agencies.
The concept behind PITT falls under the broader umbrella of predictive policing—a practice in which law enforcement agencies analyze data to identify individuals who may be worth investigating, even when no specific crime is suspected. This differs from traditional policing, which typically begins with a reported crime, a specific tip, or observed suspicious behavior.
Predictive policing programs generally rely on pattern analysis. Analysts look for combinations of data points that experience suggests may correlate with criminal activity. The theory is that identifying these patterns early could allow law enforcement to intervene before crimes occur or to focus resources more efficiently.
However, predictive policing has been controversial because it can generate suspicion based on statistical associations rather than specific evidence of wrongdoing. Civil-liberties organizations have raised concerns that such programs can lead to stops and searches that would not otherwise occur, potentially violating Fourth Amendment protections.
It is important to note that the available evidence does not establish that PITT uses artificial intelligence or machine learning systems. The reporting describes human analysts reviewing databases and making assessments, though the full scope of the program’s technical capabilities has not been publicly disclosed.
According to the 404 Media investigation, PITT personnel are assigned to review “law enforcement-sensitive databases” and develop intelligence reports that are then shared with local law enforcement agencies.
A Department of Homeland Security document obtained during the discovery phase of a criminal case describes the PITT role in detail. The document, written by Border Patrol Agent Matthew Phelps, states that his assignment involves reviewing these databases to develop intelligence before “handing over information to local law enforcement.”
The investigation reported that the intelligence developed by PITT specifically involves reviewing Americans’ financial activity. In one documented case, the analyst reported observing “information contained within law-enforcement-sensitive systems suggesting financial activity patterns commonly associated with illicit narcotics activity.”
The document also indicates that PITT personnel reviewed the individual’s prior arrest and criminal history as part of their analysis.
After developing this intelligence, PITT personnel reportedly share it with local law enforcement agencies. The stated purpose, according to the DHS document, was to “support general interdiction efforts.”
The investigation did not establish that every PITT team performs identical activities or uses identical methods. The reporting identified specific activities in documented cases but noted that CBP has declined to disclose comprehensive details about the program’s operations across all sectors.
One of the most significant elements of the 404 Media investigation is the revelation that PITT teams reportedly analyze Americans’ financial activity as part of their intelligence-gathering process.
According to the DHS document reviewed in the investigation, Border Patrol Agent Matthew Phelps reviewed financial information and identified what he characterized as patterns “commonly associated with illicit narcotics activity.” This assessment was then shared with Montana Highway Patrol, which subsequently stopped the individual.
Several important points about the financial data remain unclear:
What financial data sources are being used? CBP declined to tell 404 Media what financial activity the agency monitors or how it obtains that information.
Is a warrant required? The investigation reported that CBP would not answer whether it obtained a warrant before analyzing the individual’s financial records.
What specific patterns trigger suspicion? The agency has not disclosed the specific financial activity patterns that its analysts consider indicative of potential criminal activity.
How is financial data obtained? The investigation did not establish the specific mechanisms by which Border Patrol accesses financial information, whether through legal process, commercial data brokers, or other means.
The reporting does not establish that PITT has access to all Americans’ bank accounts or that every financial transaction is automatically reported to Border Patrol. Rather, the investigation documents that in at least one case, financial activity was analyzed as part of the targeting process.
The investigation described a process by which intelligence developed by PITT analysts reportedly reaches local law enforcement and results in traffic stops.
According to the reporting, the process generally follows this pattern:
In the documented Montana case, Border Patrol Agent Phelps reportedly sent his assessment to Montana Highway Patrol Sergeant James Beck. The DHS document states that Phelps provided the intelligence to “support general interdiction efforts.”
Sergeant Beck subsequently stopped Kyle Olson, citing an obstructed license plate as the reason for the stop. However, during discovery in Olson’s criminal case, documents revealed that Border Patrol had flagged Olson before the stop occurred.
This process has been described by civil-liberties advocates as “parallel construction”—a technique in which investigators receive information from a sensitive source and then create a separate investigative trail that conceals how the investigation actually began.
Jake Laperruque, deputy director of the Security and Surveillance Project at the Center for Democracy & Technology, told 404 Media that “genuine probable cause cannot be synthetically generated.” He added that Border Patrol appears to be “using parallel construction to cloak the reason behind its car stops in secrecy.”
The investigation did not establish that every traffic stop following a PITT lead follows the same pattern. The documented case represents one example of how the process reportedly operates.
The 404 Media investigation identified PITT teams in at least two Border Patrol sectors.
The investigation identified a PITT team operating within the Spokane Sector Border Patrol Targeting & Intelligence Division. This sector covers the U.S. border with Canada in the Pacific Northwest region.
The DHS document reviewed in the investigation was authored by Border Patrol Agent Matthew Phelps, who is assigned to this sector’s PITT team. The document describes Phelps’s role as reviewing law-enforcement-sensitive databases and developing intelligence for local law enforcement.
The investigation also identified a PITT team in the Laredo Sector, which polices the U.S. border with Mexico in South Texas.
Information about the Laredo Sector PITT team reportedly emerged from a separate case involving Alek Schott, who was stopped by a Bexar County sheriff’s deputy in Texas in 2022. During discovery in a lawsuit that followed, Schott’s movements were reportedly tracked in a WhatsApp group that included Border Patrol agents, one of whom had a profile displaying a patch reading “Laredo Sector Tech Ops PITT.”
CBP operates 20 sectors nationwide but declined to tell 404 Media how many have PITT teams or whether additional teams exist in other sectors.
It is important to note that the investigation identified these two sectors specifically. The full nationwide scope of the PITT program remains undisclosed by CBP.
The Montana traffic stop involving Kyle William Olson provided the documentary evidence that revealed the PITT program’s existence. The following account draws on police reports, body camera footage, a DHS document, and an interview Olson gave to 404 Media.
In May 2026, Olson was driving his black Chevy SUV on Interstate 90 in Montana, traveling from California to his home in Wisconsin. He was transporting marijuana edibles, marijuana cuttings, and butane hash oil obtained from his work at a licensed cannabis farm in California, he told 404 Media.
Montana Highway Patrol Sergeant James Beck was parked along the interstate when Olson drove past. According to Beck’s police report, he noticed the vehicle’s license plate was obstructed by a tinted cover and the luggage holder. Beck pulled out and conducted a traffic stop.
Body camera footage shows Beck telling Olson that his license plate was “a little bit obstructed” and stating, “Just so you know, that’s our reason for stopping you.”
During the stop, officers demanded Olson step out of the vehicle, tested his blood, and charged him with driving under the influence. Authorities later charged Olson with possession with intent to distribute based on the marijuana found in his vehicle.
During the discovery phase of his criminal case, Olson received a Department of Homeland Security document explaining how Border Patrol had instigated his traffic stop.
The document, written by Border Patrol Agent Matthew Phelps, stated that Phelps was assigned to the “Spokane Sector Border Patrol Targeting & Intelligence Division (TID) — Predictive Intelligence Targeting Team (PITT).”
In the document, Phelps wrote that he reviews “law enforcement-sensitive databases” to develop intelligence before handing information to local law enforcement. The intelligence specifically involved reviewing financial activity, the document stated.
Phelps wrote that he observed “information contained within law-enforcement-sensitive systems suggesting financial activity patterns commonly associated with illicit narcotics activity.” He also reviewed Olson’s prior arrest and criminal history.
Phelps stated that he sent his findings to Sergeant Beck to “support general interdiction efforts,” while noting that his intelligence was not the “sole basis” for the stop.
Olson told 404 Media that he works on a licensed cannabis farm in California and was transporting the marijuana products to his home in Wisconsin. He shared the DHS document with the outlet.
U.S. Customs and Border Protection provided a statement to 404 Media addressing the PITT program. The agency’s response emphasized the lawful nature of its intelligence activities while declining to discuss operational specifics.
A CBP spokesperson stated: “U.S. Border Patrol uses intelligence-informed analysis and planning to support national security operations, allocate resources, and help identify potential threats and bad actors to public safety. These efforts are conducted consistent with applicable law, policy, privacy protections, and oversight requirements.”
The agency also explained its refusal to provide additional details: “For operational security reasons, we do not discuss specific analytical methods, data sources, targeting criteria, investigative techniques, system capabilities, or deployment details. Public disclosure of such information could compromise law enforcement operations and investigations as well as creating safety risks for agents and the public.”
CBP declined to answer specific questions from 404 Media about:
The 404 Media investigation raised significant questions that remain unanswered by CBP or other authorities. These unresolved issues include:
Newz18 does not answer these questions by speculation. They remain genuinely unknown based on available public information.
Predictive policing programs like PITT raise several concerns that civil-liberties organizations and legal experts have articulated.
False positives: Data analysis can identify patterns that appear suspicious but have innocent explanations. Individuals may be flagged based on statistical associations rather than actual evidence of wrongdoing.
Bias: Algorithms and human analysts can reproduce existing biases. The Brennan Center for Justice has noted that predictive policing systems relying on historical policing data can perpetuate disparities in enforcement.
Privacy: The analysis of financial activity and other personal data raises questions about government surveillance of Americans not suspected of crimes.
Transparency: When programs operate secretly, courts and the public cannot evaluate their accuracy or legality. Critics argue that secrecy prevents meaningful accountability.
Due process: If the actual basis for a traffic stop is concealed through “parallel construction,” defendants may be unable to challenge the constitutionality of the stop.
Probable cause: The Fourth Amendment requires probable cause for searches and seizures. Critics question whether algorithmically generated suspicion meets this standard.
It is important to emphasize that these are concerns raised by experts and advocacy organizations. Courts have not definitively ruled on the constitutionality of the PITT program specifically.
Several legal and privacy experts have commented on the PITT program in the wake of the 404 Media investigation.
Jake Laperruque, Center for Democracy & Technology: Laperruque, deputy director of the Security and Surveillance Project at CDT, told 404 Media that “genuine probable cause cannot be synthetically generated.” He argued that Border Patrol appears to be “using parallel construction to cloak the reason behind its car stops in secrecy. If we can’t meaningfully review and evaluate these systems, we can’t trust them.”
Rob Frommer, Institute for Justice: Frommer, a senior attorney at IJ, told 404 Media that combining mass surveillance with predictive policing “is a recipe for tyranny.” He argued that programs like PITT treat Americans as potential suspects rather than citizens.
The Institute for Justice represents Alek Schott in a lawsuit challenging a 2022 traffic stop in Texas that allegedly involved PITT intelligence. The lawsuit alleges that the deputy fabricated the traffic offense and violated Schott’s Fourth Amendment rights.
These expert opinions represent the views of the individuals and organizations cited. They do not constitute legal rulings or established facts about the program’s legality.
The available evidence does not establish that PITT uses artificial intelligence or machine learning systems.
The 404 Media investigation describes human analysts reviewing databases and making assessments. The DHS document authored by Border Patrol Agent Phelps describes his role as reviewing “law enforcement-sensitive databases” to develop intelligence—language that suggests human analysis rather than automated algorithmic processing.
Predictive policing can take several forms:
The investigation did not establish which of these approaches PITT employs. The term “predictive” in the program’s name refers to the general concept of using data to anticipate potential criminal activity, not necessarily to AI technology specifically.
PITT is part of a broader intelligence infrastructure within Customs and Border Protection. The agency has long used various targeting systems and risk-based analysis as part of border security operations.
An Associated Press investigation published in 2025 found that Border Patrol was using automatic license plate readers as part of a wide-ranging predictive policing program. According to that reporting, the system can identify supposedly suspicious patterns of movement, including trips occurring hundreds of miles from the border, and pass information to local law enforcement.
The AP investigation found that through one company alone, Border Patrol temporarily had access to at least 1,600 license plate readers across 22 states. Some devices were reportedly disguised inside traffic cones, barrels, and roadside equipment.
Border Patrol also operates systems like the Conveyance Monitoring and Predictive Recognition System (CMPRS), which collects license plate images and matches them against hot lists to identify travel patterns.
It is important to distinguish between PITT specifically—a reported team structure within Border Patrol’s Targeting and Intelligence Divisions—and the agency’s broader intelligence enterprise, which encompasses multiple programs and technologies.
The PITT program raises practical questions for Americans concerned about privacy and law enforcement practices.
Privacy: If Border Patrol is analyzing financial activity, Americans may have their banking patterns reviewed without their knowledge. The scope of this analysis—and what triggers suspicion—remains undisclosed.
Traffic stops: The investigation suggests that individuals may be stopped based on intelligence that is not disclosed to them or to the courts. If the actual reason for a stop is concealed, challenging its legality becomes difficult.
Information sharing: PITT intelligence is reportedly shared with local law enforcement agencies. The extent and frequency of this sharing is not publicly known.
Financial data protection: The investigation raises questions about what legal protections exist for financial records and how they apply to predictive policing programs.
Government transparency: CBP’s refusal to discuss operational details means the public cannot evaluate the program’s accuracy, effectiveness, or compliance with legal requirements.
Oversight: The specific oversight mechanisms governing PITT operations have not been publicly detailed, raising questions about accountability.
The reporting does not establish that ordinary Americans are automatically being targeted. Rather, it documents a program that analyzes data to identify individuals for potential investigation, with the full scope of that targeting remaining undisclosed.
PITT stands for: Predictive Intelligence Targeting Team
Agency connection: U.S. Border Patrol / Customs and Border Protection (CBP) / Department of Homeland Security
Purpose reported: Intelligence analysis and targeting to support “interdiction efforts”
Reported information analyzed: Law-enforcement-sensitive databases, including financial activity in at least one documented case
Known sectors: Spokane Sector (Washington) and Laredo Sector (Texas) have been identified in reporting
National scope: Not fully disclosed; CBP operates 20 sectors but would not say how many have PITT teams
CBP position: Intelligence-informed analysis is conducted consistent with applicable law, policy, privacy protections, and oversight requirements
2022: Alek Schott is stopped by a Bexar County sheriff’s deputy in Texas. Discovery in subsequent litigation reportedly reveals his movements were tracked in a WhatsApp group that included Border Patrol agents, one with a profile showing “Laredo Sector Tech Ops PITT.”
2025: An Associated Press investigation finds Border Patrol using automatic license plate readers as part of a predictive policing program.
May 2026: Kyle William Olson is stopped by Montana Highway Patrol on Interstate 90. A DHS document later reveals that Border Patrol’s PITT unit had flagged his financial activity before the stop.
September 8, 2026: 404 Media publishes its investigation revealing the existence of Predictive Intelligence Targeting Teams.
September 8-9, 2026: Multiple news outlets report on the investigation, including Gizmodo, Military.com, and Latin Times.
September 9, 2026: CBP provides a statement to 404 Media defending its intelligence practices while declining to discuss operational details.
PITT stands for Predictive Intelligence Targeting Team. These are units within U.S. Border Patrol’s Targeting and Intelligence Divisions that reportedly analyze data to develop intelligence for law enforcement purposes.
Predictive Intelligence Targeting Teams are Border Patrol units that review law-enforcement-sensitive databases to identify individuals who may warrant investigation. According to recent reporting, they analyze financial activity and other data, then share intelligence with local law enforcement.
Yes. PITT operates within U.S. Border Patrol, which is part of Customs and Border Protection, an agency within the Department of Homeland Security.
According to the 404 Media investigation, PITT analysts review “law-enforcement-sensitive databases,” including financial activity and criminal history. A DHS document described analysis of “financial activity patterns commonly associated with illicit narcotics activity.”
The investigation documented at least one case where financial activity was analyzed. However, CBP has declined to disclose the specific financial data sources it uses, whether it obtains warrants, or how broadly financial analysis is applied across the program.
PITT teams have been identified in the Spokane Sector (Washington) and Laredo Sector (Texas). CBP operates 20 sectors nationwide but would not say how many have PITT teams or whether additional teams exist.
The available evidence does not establish that PITT uses artificial intelligence. The reporting describes human analysts reviewing databases and making assessments. The full technical capabilities of the program have not been publicly disclosed.
CBP stated that “U.S. Border Patrol uses intelligence-informed analysis and planning to support national security operations, allocate resources, and help identify potential threats and bad actors to public safety.” The agency said these efforts comply with applicable law, policy, privacy protections, and oversight requirements.
The legality of predictive policing depends on the specific activities, legal authority, jurisdiction, and facts of each case. Courts have not definitively ruled on the PITT program specifically. The Fourth Amendment requires probable cause for searches and seizures, and civil-liberties organizations argue that some predictive policing practices may not meet this standard.