New and Emerging Threats from Mexican Cartels Across the Air, Cyber and Territorial Domains

On 4 October 2026, soldiers of the Mexican Army working the road that runs from El Rosario toward Agua Verde, in southern Sinaloa, recovered a cache that would have looked routine on the front lines of Ukraine and was startling in Mexico. Fifteen first person view drones lay alongside fifteen controllers, fourteen batteries, firearms, 42 improvised explosive devices and fifteen spools of fiber optic cable.[1] A week later, on the same stretch of road, troops seized another 48 drones, 64 batteries and 44 spools of cable, together with ammunition, ballistic plates and tactical uniforms.[2]

The significance of the seizure lies in the cable. A drone steered through a thin optical filament neither emits nor receives a radio signal that a jammer can disrupt, which is precisely why Russian and Ukrainian units turned to the technique once electronic warfare began knocking conventional drones out of the sky. Mexico has invested heavily in jammers, and David Mora of the International Crisis Group has described signal inhibition as the government’s basic answer to criminal drones. [3] The armed groups fighting over southern Sinaloa appear to have studied that answer and moved past it. It is a sobering thought.

The official account contains two more details, and both are worth dwelling on. Defense Secretary General Ricardo Trevilla Trejo explained that the seized drones were components intended for assembly rather than finished aircraft, which suggests workshops inside Mexico capable of building to specification.[4] He also acknowledged that the Army could not determine whether the material belonged to Los Mayos or Los Chapitos, the two factions at war in Sinaloa.[5] A state that cannot say who owns a fleet of drones immune to its own countermeasures faces something more serious than a new weapon. It faces an adversary that learns faster than it does.

The argument here is simple enough to state, if harder to prove. The El Rosario seizures are a symptom of a broader transformation. Mexican cartels are acquiring new capabilities in the air, in cyberspace and on the ground at a pace that outstrips the doctrine of the Mexican and US institutions charged with confronting them. No single technology defines the threat. Its danger lies in combination, in the growing capacity of criminal armed actors to watch, strike, steal and govern while avoiding the exposure that open combat once required. An award-winning framework for forecasting criminal adoption of AI enabled autonomous drones, discussed below, suggests that the next stage of this transformation can be anticipated rather than merely observed.

From Criminal Enterprise to Armed Actor

None of this is entirely new to students of the subject. More than two decades ago, John P. Sullivan and Robert J. Bunker described how cartels, gangs and warlords could evolve into entities that contest the state’s monopoly on force and, in their most developed form, hollow out its sovereignty from within.[6] The later literature on criminal insurgency in Mexico sharpened the point. Cartels need not seek to replace the state in order to wage war on it, because their objective is freedom of action, and freedom of action is secured by neutralizing the state’s capacity to watch, arrest and prosecute.[7]

Washington’s formal vocabulary caught up in February 2025, when the Department of State designated the Sinaloa Cartel, the Cártel de Jalisco Nueva Generación (CJNG) and several other groups as Foreign Terrorist Organizations.[8] The designations moved the problem from the realm of law enforcement into that of national security and opened legal and operational tools previously reserved for jihadist networks. Yet a designation is a photograph. It records what an organization is at the moment of listing and says little about what it is becoming.

What they are becoming is harder to name. The most useful label, for now, is capability acquisition across domains, clumsy as the phrase may be. Armed criminal groups have always adopted technology opportunistically, from encrypted phones to semisubmersibles. What has changed is speed and breadth. Battlefield innovations from Ukraine now reach the Sierra Madre within months, artificial intelligence tools built for software engineers are repurposed for financial intrusion, and improvised minefields turn rural roads in Michoacán into contested terrain. Each development matters on its own. Together they describe an adversary that no longer needs to win firefights to hold ground.

Practitioners who worked public security in Tamaulipas a decade and a half ago will remember when the frightening innovation was the improvised armored truck, the monstruo, welded together in clandestine workshops. The workshops are still there. What comes out of them has changed, and it has changed faster than most of the institutions that once seized those trucks.

Mexico offers unusually favorable conditions for this kind of acquisition. Its largest criminal organizations command substantial revenues, maintain international supply chains for precursor chemicals and benefit from a steady flow of firearms from the United States. They also operate in a country with deep commercial ties to Asian manufacturers of electronic components and with long, busy borders that make the interdiction of dual use goods difficult. Few nonstate actors anywhere combine money, logistics and access to technology on this scale, which helps explain why innovations that remain marginal elsewhere tend to mature quickly in Mexico.

Anticipating where that adversary goes next requires more than cataloguing incidents. It requires a way to forecast which capabilities criminal organizations will adopt, when and under what conditions. The framework for predicting the adoption of AI enabled autonomous drone capabilities by transnational organized crime and FTOs, led by the author with Amanda Davies and Simón Levy, was designed for that purpose. Presented at the European, Mediterranean and Middle Eastern Conference on Information Systems in Paris in 2026, it received the conference’s Best Completed Research Paper Award, a recognition that reflects growing scholarly attention to the intersection of artificial intelligence, unmanned systems and organized crime.[9]

The framework’s central premise is that criminal adoption of a capability follows observable conditions rather than chance, and that those conditions can be identified before the capability appears in the field. That premise shifts the analytical task from reacting to the latest seizure toward anticipating the next one. It also gives this essay its organizing logic. The sections that follow apply it to three domains, the air, cyberspace and the ground itself, before turning to what the pattern means for a security relationship between Mexico and the United States that still divides the problem between police officers and soldiers.

The Air Domain

Cartel use of drones in Mexico has evolved through recognizable stages. Commercial quadcopters first served as eyes, watching checkpoints, rivals and military convoys. The same platforms soon began carrying explosive charges released over enemy positions, and more recently first person view kamikaze drones, flown directly into their targets by operators wearing goggles, have become a familiar weapon in Michoacán, Jalisco and Sinaloa. Fiber optic guidance marks the latest stage, and the El Rosario seizures show that it has moved from isolated videos into organized procurement. Mora notes that such drones had been flying in Mexico for months before the October seizures, first glimpsed only in scattered footage of single aircraft.[10]

Each stage has arrived faster than the one before it, and the reason is not difficult to locate. In July 2025, Defense News reported that Ukraine’s Security Service had opened a joint investigation with military intelligence after Mexico’s National Intelligence Center warned that Mexican volunteers had joined foreign fighter units specifically to acquire first person view drone skills. The inquiry later expanded to Colombian nationals.[11] Just over a year later, a US assessment reported by the Washington Post described the transmission mechanism in more concrete terms. According to a Mexican official, Colombian drone specialists returning from Ukraine are offered large sums and are sometimes brought to Mexico to train groups of cartel droneros.[12] Not every account agrees, to be fair. One Colombian fighter in Ukraine told the newspaper that he had never heard of Mexican cartels infiltrating the ranks.[13] Still, the evidence points toward a channel through which the most intensive drone war in history is exporting its lessons to Latin American criminal organizations, with Colombian veterans as the main carriers.

Autonomy is the logical next step. Fiber optic guidance solves the jamming problem at a price, because the cable limits range, adds weight and can snag on terrain. Autonomous terminal guidance offers another solution, allowing a drone to complete its final approach even after losing its link to the operator. This is precisely the transition that the award-winning adoption framework discussed above was built to forecast.[14] The El Rosario workshops suggest that the industrial base for that step is already forming. Groups able to assemble fiber optic drones from imported components are within reach of autonomous terminal guidance, which requires software rather than new hardware.

Incentives push the same way. Every improvement in state countermeasures raises the value of the next adaptation, and Mexico’s reliance on jamming created the very pressure that fiber optic guidance relieves. Components are commercially available, inexpensive and easy to import in pieces that attract little attention at customs. Trained operators can be hired rather than developed, and the war in Ukraine continues to produce them in large numbers. Where a capability is useful, accessible and supported by a supply of skilled labor, adoption tends to follow. None of these conditions is likely to weaken soon, and autonomy will probably reach Mexico through the same channels, as open source flight software matures and as veterans who have used it in Ukraine find buyers for their skills.

Mexico’s own response has gone in a different direction. President Sheinbaum has announced that the Navy already operates a drone production plant and that the Secretariat of National Defense is developing its own model, while the Secretariat of Science presented a surveillance prototype called Quetzal. [15] These programs strengthen the state’s ability to see. They do little against a threat that has moved outside the radio spectrum, where jammers, until recently the cornerstone of Mexican drone countermeasures, no longer reach. Anyone who has sat through procurement meetings in a Mexican security agency will recognize the pattern, equipment bought against yesterday’s threat and delivered just in time for tomorrow’s. The US assessment reported by the Washington Post was blunter, warning that both countries are poorly prepared for what the shift in cartel drone capability will demand.[16]

The Cyber Domain

The clearest evidence that Mexican cartels will turn digital tools against the state comes from US government records. In June 2025, the Office of the Inspector General of the US Department of Justice published an audit of the FBI’s efforts to counter what it calls ubiquitous technical surveillance. The report describes how, in 2018, a hacker working for the Sinaloa Cartel during the investigation of Joaquín “El Chapo” Guzmán identified the FBI’s Assistant Legal Attaché at the US Embassy in Mexico City. Using the official’s mobile number, the hacker obtained call records and geolocation data, then exploited Mexico City’s camera network to follow the attaché and identify the people the official met. [17] According to the case agent cited in the report, the cartel used that information to intimidate and, in some cases, kill potential sources and cooperating witnesses.[18]

Why dwell on a case now eight years old? Because it shows intent, since a cartel deliberately hunted a US federal officer through the official’s digital trail. It also shows method, since the attack exploited the surveillance infrastructure of a Mexican city rather than any exotic tool. The cartel did not need to build a cyber capability from scratch. It hired one and pointed it at the people most dangerous to its survival.

The vulnerability it exposed has only grown. Mexican cities have expanded their command and control centers, camera networks and related data systems over the past decade, and each expansion enlarges the surface that a criminal organization can penetrate. A cartel with access to those systems sees what the state sees, follows those who follow it and identifies informants before they can testify. Surveillance infrastructure built to protect citizens becomes, once compromised, an instrument of counterintelligence for the very groups it was meant to monitor.

Today the capability on offer is of a different kind altogether. On 21 October 2026, researchers at the security firm Straiker published an investigation into a Mexican criminal group operating as Tr4nsHack Team under the front company Hexagon Security Labs Mexico. According to the report, the group built a system called CARLIY0S in which several commercial AI models, including Anthropic’s Claude, xAI’s Grok and Moonshot AI’s Kimi, were assigned distinct roles in reconnaissance, exploit development and decision making against more than 400 banks and payment providers across Latin America.[19] A snapshot of the group’s stolen data contained 4,493 records, including 2,911 email addresses and 903 account numbers, alongside fintech credentials and API keys that Straiker assessed could enable abuse of SPEI, Mexico’s interbank payment system. The operators persuaded the models to cooperate by presenting their requests as authorized penetration testing or academic exercises. 20]

Straiker attributes the operation to operators from CyberCartel, a financially motivated actor tracked since 2012, which it describes as having presumed ties to Mexican cartels and a division of labor that extends to money mule networks for laundering stolen funds.[21] The report offers no direct evidence of that cartel link, and it should not be overstated. A separate wave of AI assisted intrusions against Mexican government bodies, beginning with the tax authority in December 2025, likewise remains unattributed to any drug trafficking organization.[22]

The FTO designations add a further incentive. Terrorism designations expose banks, suppliers and intermediaries that handle cartel money to material support liability and sanctions, which raises the cost of laundering through conventional channels. An organization under that pressure has strong reasons to seek alternatives, from cryptocurrency to direct manipulation of payment systems. Groups able to harvest credentials for interbank transfers, maintain persistent access inside financial institutions and move stolen funds through mule networks offer the very services a sanctioned cartel would need. The designation intended to isolate cartels financially may therefore accelerate their interest in digital capabilities. None of this argues against designation, but it does show that every instrument of pressure generates adaptation, and that the adaptation must be anticipated rather than discovered after the fact.

Attribution, in the end, matters less than it seems. Mexico now hosts criminal actors able to orchestrate AI driven attacks on the financial system at a scale no human team could match, and Mexico’s largest cartels have already shown that they will purchase cyber capability and use it against the state when their survival is at stake. Convergence between the two is therefore a matter of acquisition rather than invention, the dynamic that adoption frameworks are designed to anticipate. A cartel that combines this kind of financial intrusion with the surveillance tradecraft of 2018 would hold a capability once reserved for intelligence services.

The Territorial Domain

Drones and malware are means. The end is territory, and it is on the ground that their purpose becomes clear. On 3 July 2026, speaking at the presidential morning press conference in Morelia, General Trevilla reported that specialized Army personnel had detected 625 improvised explosive devices in Michoacán since November 2025, a mix of landmines and explosives delivered by drone.[23] Demining operations were concentrated on the routes from El Guayabo toward El Alcalde, in the Tierra Caliente region around Apatzingán, and the Secretariat announced a project to build mine resistant vehicles and acquire drones capable of detecting buried charges.[24]

The general’s explanation of who was laying the mines is as revealing as the number. He attributed them to a confrontation in which the CJNG seeks to enter Tepalcatepec and Coalcomán while consolidating its hold around Apatzingán, and in which local groups place mines to defend those areas from the incursion.[25] The minefield, in other words, has ceased to be a CJNG monopoly and become a tool of territorial denial for smaller organizations seeking to hold ground against a stronger rival, with consequences for anyone who travels those roads, soldiers and residents included. The pattern was visible earlier. In March 2024, six simultaneous detonations led Michoacán’s state security secretariat to create its own bomb disposal unit.[26] What was then an alarming anomaly is now an operational routine.

The common thread is remote violence. Mines, explosive drones and fiber optic strikes allow armed groups to kill, deter and deny terrain without exposing their fighters to the firepower of the armed forces. This is the logic of insurgency rather than the logic of a gunfight, and it serves the same end that criminal insurgency theory identified long ago, namely freedom of action within a territory the state cannot fully enter. That freedom is then converted into revenue and authority through extortion, control of local economies and influence over local institutions. When President Sheinbaum told the same Morelia conference that eradicating extortion was the main objective of the federal strategy in Michoacán, she was describing a form of parallel governance rather than an ordinary crime. The same conference reported that some 10,000 soldiers and National Guard personnel were deployed in the state.[27]

Remote violence also reverses the cost structure of the conflict. A landmine or an explosive drone costs a small fraction of what the state must spend to detect it, clear it and protect its forces against it. The daily consequences, familiar to anyone who has commanded patrols in contested municipalities, are mundane and exhausting. Convoys travel only in daylight, routes are varied, drivers slow down at every culvert. Each precaution is a small victory for whoever planted the device. The announced program of mine resistant vehicles and detection drones is a rational response, but it is also an admission that the armed forces must now plan for a battlefield shaped by their adversaries. In Tierra Caliente, the state is no longer simply pursuing criminals. It is operating on ground that criminals have prepared.

The resilience of this model was tested in February 2026, when Mexican forces killed Nemesio Oseguera Cervantes, “El Mencho,” the founder of the CJNG. In an October 2026 assessment, the Overseas Security Advisory Council of the US Department of State concluded that large scale fragmentation of the organization had not occurred and that the CJNG brand may have outgrown loyalty to its founder.[28] It also noted that residents of Michoacán reported a resumption of CJNG drone attacks in March, only weeks after his death. [29] Leadership decapitation, the central instrument of two decades of counternarcotics policy, struck the most wanted man in Mexico and left the organization’s capabilities largely intact. Meanwhile, the war between Los Chapitos and Los Mayos, which began in Sinaloa in September 2024, has become the most likely destination for the fiber optic drones seized at El Rosario.

Implications for Mexico and the United States

These three domains do not simply happen to overlap. Drones extend the reach of territorial control, cyber capability protects and finances the enterprise while blinding those who pursue it, and remote violence converts both into durable authority over people and places. An organization that combines them is neither a criminal enterprise with an armed wing nor an insurgency with a business model. It is a criminal armed actor operating across several domains at once, and it is adapting faster than the institutions designed to contain it.

There are limits to this argument, and they should be stated plainly. Firearms remain the instrument of most homicides in Mexico, and the capabilities described here are concentrated in a few regions and organizations rather than spread evenly across the criminal landscape. Attribution is often incomplete, as the Army’s uncertainty about the El Rosario drones and Straiker’s caution about CyberCartel both show. Fiber optic drones also carry real constraints of range and maneuver. None of these qualifications weakens the central claim. The question for policy is not whether every cartel has become an actor operating across several domains, but whether the most capable ones are moving in that direction faster than the state is adapting. The evidence suggests that they are.

Countermeasures are the most immediate concern. Mexico’s heavy investment in jamming has been overtaken quickly, and the next generation of cartel drones may not depend on a continuous link at all. Detection, physical interception and the protection of convoys and bases must take priority over spectrum denial, and they should be developed jointly with US Northern Command and with Colombian forces, which face a similar drone threat from armed groups at home.

Equally pressing are the people behind the machines. The main channel for cartel drone expertise appears to run through foreign volunteers returning from Ukraine. Intelligence sharing among Mexico, Colombia, Ukraine and the United States on the recruitment of returning fighters by criminal organizations would address the threat at its source, at a fraction of the cost of defending against its products.

The financial system presents a different kind of exposure. The CARLIY0S case shows that AI driven intrusion against banks and payment rails in Mexico is no longer theoretical. Regulators, the central bank and the private sector need threat sharing arrangements that treat AI orchestrated campaigns as a national security problem, because the line between the hackers who build these systems and the cartels that would rent them is increasingly porous.

The state’s own surveillance systems deserve similar attention. The 2018 case shows that municipal camera networks and telecommunications data can be turned against investigators and witnesses. Hardening those systems, auditing who can access them and treating their compromise as a counterintelligence failure rather than a technical incident should form part of any binational agenda.

Beneath all of these lies a doctrinal problem. The relationship between Mexico and the United States still tends to divide the cartel problem between police officers who arrest traffickers and soldiers who fight armed groups. The organizations described in this essay fit neither category. They require a response that combines intelligence, financial enforcement, military force and the reconstruction of local governance, sustained over years rather than measured in the number of leaders captured or killed.

And success will have to be measured differently. Homicide figures, arrests and the number of leaders removed capture only part of a conflict in which armed groups increasingly kill at a distance, deny territory with explosives and extract revenue through extortion. Indicators such as explosive device detections, drone incidents, reported extortion and the ability of public officials to operate freely in contested municipalities would give policymakers on both sides of the border a more accurate picture of who actually controls the ground.

A few dozen spools of fiber optic cable, recovered on a back road near El Rosario, will not change the balance of power in Sinaloa, or anywhere else, on their own. Their significance lies in what they reveal. Mexican cartels are learning from the most advanced battlefield of the decade, buying capabilities that once belonged to intelligence services and using them to govern territory the state cannot reach. The threat is no longer emerging. It has arrived, and the institutions charged with confronting it are still catching up.

Endnotes

[1] “Drones de fibra óptica: la tecnología de la guerra de Ucrania y Rusia que llegó a los cárteles mexicanos.” Infobae. 18 October 2026, https://www.infobae.com/mexico/2026/10/18/drones-de-fibra-optica-la-tecnologia-de-la-guerra-de-ucrania-y-rusia-que-llego-a-los-carteles-mexicanos/.

[2] Xavier Jiménez, “Aseguran otros 48 drones para volar con fibra óptica en Sinaloa.” Milenio. 12 October 2026, https://www.milenio.com/policia/decomisan-48-drones-volar-fibra-optica-en-sinaloa.

[3] “La incautación de drones de fibra óptica en Sinaloa genera preocupación en México.” Texas Public Radio. 21 October 2026, https://www.tpr.org/tpr-noticias/2026-10-21/la-incautacion-de-drones-de-fibra-optica-en-sinaloa-genera-preocupacion-en-mexico.

[4] “Preocupa detección de material de drones manejados con fibra óptica en México: Defensa.” Milenio. 13 October 2026, https://www.milenio.com/politica/preocupa-deteccion-material-drones-fibra-optica-mexico-defensa.

[5] Ibid.

[6] John P. Sullivan and Robert J. Bunker, “Drug Cartels, Street Gangs, and Warlords.” Small Wars & Insurgencies. Vol. 13, no. 2, 2002: pp. 40–53, https://doi.org/10.1080/09592310208559180.

[7] Robert J. Bunker, Ed., Criminal Insurgencies in Mexico and the Americas: The Gangs and Cartels Wage War. London: Routledge, 2013; John P. Sullivan and Robert J. Bunker, Eds., The Rise of the Narcostate (Mafia States). Bloomington: Xlibris, 2018; John P. Sullivan. “Crime wars: Operational perspectives on criminal armed groups in Mexico and Brazil.” International Review of the Red Cross. IRRC No, 923, June 2023, https://international-review.icrc.org/articles/crime-wars-operational-perspectives-923.

[8] US Department of State, “Foreign Terrorist Organization Designations of Tren de Aragua, Mara Salvatrucha, Cartel de Sinaloa, Cártel de Jalisco Nueva Generación, Carteles Unidos, Cartel del Noreste, Cártel del Golfo, and La Nueva Familia Michoacana.” Federal Register. Vol. 90, no. 33. 20 February 2025,https://www.federalregister.gov/documents/2025/02/20/2025-02873/foreign-terrorist-organization-designations-of-tren-de-aragua-mara-salvatrucha-cartel-de-sinaloa.

[9] Ghaleb Krame, Amanda Davies and Simón Levy, “A Framework for Predicting the Adoption of AI-Enabled Autonomous Drone Capabilities by Transnational Organized Crime and FTOs.” Proceedings of the European, Mediterranean and Middle Eastern Conference on Information Systems (EMCIS 2026). Paris, 2026, forthcoming.

[10] Op. cit., Texas Public Radio at Note 3. See also, Robert J. Bunker and John P. Sullivan, Eds., Criminal Drone Evolution: Cartel Weaponization of Aerial IEDs. Bloomington: Xlibris, 2021; Robert J. Bunker, “Three Cartel FPV Racing Drones Recovered in Concordia, Sinaloa: Fiber Optic C2 Evident.” C/O Futures Mexican Cartel Imagery Archive. 9 October 2026, https://www.cofutures.net/post/three-cartel-fpv-racing-drones-recovered-in-concordia-sinaloa-fiber-optic-c2-evident.

[11] “Drug Cartel Operatives Snuck into Ukraine for Drone Training, Report.” Defense News. 30 July 2025, https://defensenews.com/global/the-americas/2025/07/30/drug-cartel-operatives-snuck-into-ukraine-for-drone-training-report.

[12] Alex Horton, Samantha Schmidt and Warren P. Strobel “Drug Cartels Are Making Alarming Drone Advances, US Assessment Warns.” Anchorage Daily News. 22 September 2026, https://www.adn.com/nation-world/2026/09/22/drug-cartels-are-making-alarming-drone-advances-us-assessment-warns/

[13] Ibid.

[14] Op. cit., Krame, Davies and Levy at Note 9.

[15] Op. cit., Texas Public Radio at Note 3.

[16] Op. cit., Anchorage Daily News at Note 12.

[17]“Audit of the Federal Bureau of Investigation’s Efforts to Mitigate the Effects of Ubiquitous Technical Surveillance.” Audit Report 25 065. Washington, DC: US Department of Justice, Office of the Inspector General. June 2025, https://oig.justice.gov/sites/default/files/reports/25-065_t.pdf.

[18] Ibid.

[19] STAR Labs Research, “Inside CARLIY0S: How CyberCartel Used Multiple AI Models to Attack 400+ Latin American Banks and Financial Institutions.” Straiker. 21 October 2026, https://www.straiker.ai/blog/inside-carliy0s-how-cybercartel-used-multiple-ai-models-to-attack-latin-american-banks.

[20] Ibid.

[21] Ibid.

[22] Ionut Arghire, “Hackers Weaponize Claude Code in Mexican Government Cyberattack.” SecurityWeek. 1 March 2026, https://www.securityweek.com/hackers-weaponize-claude-code-in-mexican-government-cyberattack/.

[23] Ricardo Trevilla Trejo, remarks at the morning press conference of President Claudia Sheinbaum Pardo, Morelia, Michoacán. Gobierno de México. 3 July 2026, video, https://www.youtube.com/watch?v=KBbSB8MWwPg.

[24] Nestor Jiménez, “Eliminar la extorsión en Michoacán, principal objetivo de Sheinbaum.” La Jornada. 4 July 2026, https://www.jornada.com.mx/2026/07/04/estados/022n1est’.

[25] Miguel García Tinoco, “Más de 600 artefactos explosivos han sido desactivados en Michoacán, son colocados por disputas delincuenciales: Trevilla.” Excélsior. 3 July 2026, https://www.excelsior.com.mx/nacional/explosivos-desactivados-michoacan.

[26] Miguel Flores, “Narcominas en Michoacán son producto de enfrentamientos entre CJNG y cárteles locales, asegura Trevilla Trejo.” Infobae. 3 July 2026, https://www.infobae.com/mexico/2026/07/03/narcominas-en-michoacan-son-producto-de-enfrentamientos-entre-cjng-y-carteles-locales-asegura-trevilla-trejo/.

[27] Op. cit., La Jornada at Note 24.

[28] Ghaleb Krame, “El Mencho Is Dead. The Brand Isn’t,” Ghaleb Krame, Ph.D. on Substack, 20 October 201d, https://gkrame.substack.com/p/el-mencho-is-dead-the-brand-isnt referencing  “CJNG After El Mencho.” Washington, DC: Overseas Security Advisory Council. October 2026.

[29] Ibid.