FBI Probes Suspected Breach of National Crime Database as Hackers Hit USDA Farm Payment System
The Federal Bureau of Investigation is investigating a suspected intrusion into its National Crime Information Center, the vast database that police officers across the United States rely on to check warrants, wanted persons and stolen property, according to CNN, which first reported the incident on Friday. In the same week, hackers breached a system at the United States Department of Agriculture that sends payments to farmers, NBC News reported, citing two former officials with knowledge of the matter.
The two incidents, disclosed within hours of each other on Friday 20 November, struck at the heart of two very different but equally critical federal systems: one that underpins everyday law enforcement across the country, and one that moves money to the farms that feed it. In neither case have officials said who was responsible, how the attackers got in, or whether any data was stolen or altered.
Two people briefed on the FBI investigation told CNN that officials discovered the suspected intrusion into the National Crime Information Center, known as NCIC, in recent days. The bureau has not disclosed how the attackers gained access or whether they obtained confidential records, and it did not respond to CNN’s request for comment at the time of the report, according to the network.
At the Agriculture Department, hackers gained access to the system that facilitates payments to farmers, and the department told employees about the incident on Thursday 19 November, according to the two former USDA officials cited by NBC News. Reuters carried the NBC report on Friday. The Cybersecurity and Infrastructure Security Agency, the federal government’s main civilian cyber defence agency, referred questions to the USDA, which did not immediately respond to a request for comment, Reuters reported.
Neither agency has named a suspect, and no group has claimed responsibility for either attack. That silence, combined with the sensitivity of the systems involved, has left two central questions unanswered: what the intruders saw, and whether what they saw can still be trusted.
What Happened at the FBI
The NCIC is not a peripheral database. It is the electronic clearinghouse of American criminal justice information, available to virtually every law enforcement agency in the country, 24 hours a day, 365 days a year. It was launched on 27 January 1967, and today it holds around 12 million active records across 21 files, handling an average of about 14 million transactions a day, according to the FBI. More than 90,000 criminal justice and law enforcement agencies use it, along with judges, prosecutors, corrections officers and court administrators.
When a police officer runs a licence plate during a traffic stop, the query is compared against property files covering stolen vehicles, guns, boats, licence plates, securities and articles, and against person files that include wanted persons, missing persons, sex offenders, protection orders, gang members, immigration violators and suspected terrorists. A wanted-person hit is confirmed in NCIC roughly every 90 seconds, according to the FBI. The system is housed at the bureau’s Criminal Justice Information Services Division facility in Clarksburg, West Virginia, which also holds the FBI’s fingerprint database and the National Instant Criminal Background Check System used for gun buyers.
That scale is precisely why the suspected intrusion matters. A former FBI official who previously investigated cybercrime groups told CNN that data integrity presents a major concern. Hackers accessing sensitive records would already pose a significant threat, the former official said, but manipulating information could create additional risks for officers and civilians. Inaccurate records could lead officers to rely on incorrect information during law enforcement operations, and investigators must now determine whether the suspected attackers simply accessed the database or gained deeper control over it.
The FBI has not confirmed whether the intrusion affected the accuracy of NCIC records. No alteration of records has been confirmed, and the bureau has not said whether any confidential records were obtained. The distinction matters enormously: a breach in which data is merely viewed is a serious security failure, but a breach in which records are quietly changed could put wrong information in front of an officer approaching a car, or erase a warning that should have been there.
The suspected intrusion follows several other security incidents involving federal law enforcement systems. In September, the ShinyHunters hacking group targeted sensitive employee information through an FBI recruitment website. Some FBI personnel reportedly criticised the agency’s communication and security support following that incident, and a Canadian cybersecurity researcher and author now faces charges connected to the ShinyHunters investigation. An earlier incident in 2023 involved a computer system at the FBI’s New York field office. Hackers breached infrastructure used in child sexual exploitation investigations, and the affected systems also included storage connected to the bureau’s Jeffrey Epstein investigation.
The USDA Breach and the Farmer Payment System
The Agriculture Department incident unfolded over a longer stretch, and its operational effects were felt well before the public learned of it. Nextgov, a publication covering federal government information technology, first reported on Thursday that the department’s Cybersecurity and Privacy Operations Center had uncovered a security incident. An internal email sent on Monday by Chris Rottler, acting assistant chief information officer for the Farm Production and Conservation Business Center, said teams had to reset all database passwords and restart applications.
The outages hit the Farm Service Agency’s loan software and conservation-payment programmes, along with mapping and contracting systems, according to DTN, an agriculture trade publication, and ABC News. DTN also reported delays in farm loans, subsidy payments and farmland conservation work. According to Nextgov, the department suspended access to several business systems while it reset database passwords and restarted related services. As of late Thursday, outages were still intermittent and some staff could not even connect to the virtual private network used to work remotely.
The timing could hardly have been worse for farmers. The outage dragged on for more than a week as producers tried to finish enrolment in the Agriculture Risk Coverage and Price Loss Coverage programmes for the 2026 crop year, with a deadline of 11 December. Financial tools used to move money between the department and farmers, including those tied to crop insurance and disaster relief, were disrupted, according to Nextgov and AgWeb.
While the payments database is the only known extent of the intrusion so far, the former officials who described the incident to NBC News warned that the compromised system could provide a backdoor to more sensitive information, including food safety data and research on diseases that spread from animals to humans. It has not been confirmed that such research data was compromised.
The department says no data was lost. Richard Fordyce, under secretary of agriculture for farm production and conservation, acknowledged in remarks at a press briefing reported by DTN that the incident was a cyber threat rather than a simple software update, while saying security systems worked and no personal information or data was lost. He said recovery work was under way and that most systems should be back by the close of business on Friday. NBC News described the incident as the first known breach at the department initiated by an outside actor.
The Agriculture Department is not new to the list of federal agencies hit by cyber intrusions. It was among the agencies reportedly affected by the 2020 federal government data breach, a supply-chain compromise acknowledged in December 2020 that reached across the US government. The current incident, however, appears to be a direct intrusion into the department’s own systems rather than a compromise delivered through a third-party supplier.
What Is at Stake
The two incidents are different in kind, but they share a common thread: both targeted systems that ordinary Americans depend on without ever seeing them. A farmer waiting on a loan or a subsidy payment does not think about the database that moves the money. A police officer approaching a vehicle at night does not think about the database that says whether the person inside is wanted. Both systems work precisely because they are invisible, and both become dangerous the moment their reliability is in doubt.
At the FBI, the immediate risk is not only that criminal records may have been viewed by unauthorised people. It is that the records may have been changed. The NCIC works on a shared management concept: the FBI provides the host computer and the network, while the agencies that enter records are responsible for entering, modifying and removing them. The bureau’s Criminal Justice Information Services Division serves as custodian of the records. That structure means an intrusion at the centre could, in principle, touch records entered by thousands of separate agencies, and verifying the integrity of 12 million records is a very different task from confirming that a single server was accessed.
At the USDA, the risk runs in the other direction. The payment system is the department’s front door to the farm economy, and the former officials’ warning that it could serve as a gateway to food safety data and animal-disease research points to the deeper concern: an attacker inside a payment database may be able to move sideways into systems that hold information about the food supply and about diseases that cross from animals to humans. None of that has been confirmed as compromised, but the fact that former officials raised it publicly suggests the possibility is being taken seriously inside the department.
The operational disruption also matters on its own terms. Farm loans, subsidy payments and conservation work were delayed at a moment when producers were racing a December enrolment deadline. For a farm business, a delayed payment is not an inconvenience; it is a missed payroll, a postponed purchase, a loan that cannot be serviced. The department has said most systems should be restored by the end of the week, but the outage had already lasted more than a week by the time it was publicly reported.
A Pattern of Attacks on Federal Systems
The two incidents did not arrive out of nowhere. They are the latest in a year of cyberattacks that have repeatedly found their way into American government and infrastructure systems, and they follow a warning from the FBI itself about how fast such attacks have become.
On 13 November, the bureau issued a cybersecurity advisory about ARTEX, an open-source penetration testing framework that malicious actors have repurposed. The framework uses large language model-powered agents to automate different stages of cyber intrusions, including reconnaissance, vulnerability exploitation and data extraction. According to the advisory, some attacks progressed from initial targeting to privileged access in under three hours, and researchers had identified more than 1,600 publicly exposed ARTEX console instances by early November. The FBI warned that attackers could use these capabilities against government agencies and other critical sectors, and recommended restricting exposed applications, strengthening access controls and monitoring suspicious account activity. The advisory does not identify ARTEX as the tool behind the suspected NCIC intrusion, and investigators have yet to establish how attackers reached the criminal information network or what information they may have compromised.
The pattern extends beyond Washington. In August, fridges at a small number of US military bases were affected in incidents under investigation as suspected cyberattacks, prompting a joint advisory from CISA and the FBI on 5 November. In October, hundreds of supermarkets and food processing sites in France were hit in attacks targeting refrigeration systems, with around 700 commercial sites affected and 15 fires reported. In early November, dozens of supermarkets across seven European countries were struck. The targets differ, but the lesson is the same: the systems that keep food cold, money moving and police informed are all connected, and all of them are being probed.
Against that backdrop, FBI Director Kash Patel used a Fox News interview to highlight the Trump administration’s record, describing Operation Winter Shield, an effort to push companies to strengthen security. “Previous administrations could have improved the systems but did not,” he said. “We are staying ahead of the threat and making up for lost time.” CNN reported that the director’s television appearance came after the security incident was known inside the FBI, and noted that he described Winter Shield, which ended months earlier, as if it were still under way. The FBI’s Chicago field office had already announced the results of the two-month Winter Shield operation in April.
Analysis: Why It Matters
The most striking feature of this story is not what the hackers did, but what the government has not said about it. In both cases, the public learned of the incidents from news organisations citing people familiar with the matter, not from the agencies themselves. The FBI did not respond to CNN’s request for comment. The USDA did not immediately respond to Reuters. CISA, the agency created to defend civilian federal networks, referred questions to the USDA. The department’s own account of the incident reached its employees in an internal email and reached the public through former officials speaking to NBC News.
That is a communications failure as much as a security failure, and it has practical consequences. When an agency will not say what happened, the public fills the gap with speculation, and the people most affected, farmers waiting on payments and police officers relying on NCIC, are left to guess whether the systems they depend on can be trusted. The USDA’s under secretary did eventually speak, and his assurance that no data was lost is welcome, but it came after days of outage and only after the story had broken in the press. The FBI has said nothing at all in public about the NCIC intrusion beyond what two briefed people told CNN.
There is a second, deeper problem. Both incidents expose how much of American public life now runs through a small number of centralised databases. The NCIC is a single point of failure for law enforcement information across the entire country: if its records cannot be trusted, every warrant check, every missing-person search and every stolen-vehicle query in America is degraded at once. The USDA payment system is a single point of failure for farm finance: if it goes down, loans, subsidies and conservation payments stop together. Centralisation brings efficiency, but it also concentrates risk, and the events of this week show what happens when that risk materialises in two places at once.
The timing also deserves attention. The USDA incident was discovered more than a week before it was publicly reported, and the NCIC intrusion was discovered “in recent days” before CNN reported it on Friday. In both cases, the agencies had time to assess the damage before the public knew anything had happened, and in both cases the public account is still incomplete. That is understandable during an active investigation, but it is also a reminder that the first version of events in a cyber incident is almost never the full version. The history of major breaches, from the 2020 supply-chain compromise that reached the USDA itself to the intrusions disclosed this week, is a history of initial reports that understated the scope, followed by months of revision.
The data-integrity question at the FBI is the one to watch most closely. Theft of data is damaging, but alteration of data is corrosive in a different way. A stolen record can be contained; a changed record can mislead. If a wanted-person entry is deleted, a dangerous person walks free. If a protection order is altered, a victim loses protection. If a stolen-vehicle record is changed, an innocent driver is stopped at gunpoint. The FBI has not confirmed any alteration, and it would be wrong to assume the worst, but the fact that a former FBI cybercrime official raised the possibility unprompted tells us that investigators are treating it as a live question, not a hypothetical one.
At the USDA, the question to watch is different: whether the intrusion stopped at the payment system or reached further. The former officials’ warning about food safety data and animal-disease research is not a casual aside; it is the kind of warning that officials give when they believe the attacker’s access was broader than the agency has admitted. If the payment system was indeed a gateway, the investigation will need to establish exactly how far the intruders travelled inside the department’s networks, and the department will need to say so publicly when it knows.
There is also a political dimension that cannot be ignored. The FBI director’s decision to appear on television to promote a concluded security operation, apparently after the NCIC incident was known inside the bureau, invites an obvious question: why was the public told about an operation that ended in April, but not about a suspected intrusion into the country’s most important crime database? The answer may be innocent, a matter of scheduling and classification, but the optics are poor, and they feed a broader scepticism about whether federal agencies are candid about their own vulnerabilities. Trust in these systems is not an abstract good; it is the thing that makes a police officer act on a warrant hit and a farmer plan around a payment date. Once that trust is shaken, it is hard to rebuild, and silence is the worst way to rebuild it.
Finally, the two incidents should be read together with the FBI’s own warning about ARTEX. The bureau’s advisory described attacks that move from initial targeting to privileged access in under three hours, driven by automated agents. That is a world in which the defenders have hours, not weeks, to detect and contain an intrusion. The USDA incident, by contrast, appears to have gone undetected or at least unreported for more than a week. The gap between the speed of the attack and the speed of the response is the central challenge of modern cybersecurity, and it is a gap that no amount of reassurance can close. Only detection, disclosure and repair can do that, and on this week’s evidence, all three are still works in progress.
What to Watch Next
Several things will determine how serious these incidents turn out to be. First, whether the FBI confirms or rules out any alteration of NCIC records, and whether it says how the attackers gained access. A confirmed intrusion with no data altered is a very different story from one in which records were changed, and the bureau’s silence on that point is the single most important gap in the public record.
Second, whether the USDA completes its restoration of the affected systems and says plainly whether the intrusion reached beyond the payment database. Farmers will also be watching the 11 December enrolment deadline for the Agriculture Risk Coverage and Price Loss Coverage programmes, and whether the disruption causes anyone to miss it.
Third, whether either agency attributes the attacks. No suspect has been named and no group has claimed responsibility, and attribution in cyber incidents often takes weeks or months, if it comes at all. But attribution matters: a criminal gang seeking money behaves differently from a state actor seeking leverage, and the response, from sanctions to indictments, depends on knowing which one it was.
Fourth, whether Congress takes an interest. Both the FBI and the USDA answer to congressional oversight, and a suspected breach of the national crime database is precisely the kind of incident that prompts hearings. The questions will be familiar ones: when was the intrusion discovered, when was it disclosed, what was done to contain it, and what will change to prevent the next one.
And fifth, whether the pattern continues. Two federal agencies in one week, following a year of attacks on military, commercial and food systems, suggests a threat environment that is not abating. The FBI’s own advisory warned that automated attacks are accelerating. The test of the government’s response will not be whether it can prevent every intrusion, because no one can, but whether it can detect them quickly, disclose them honestly and repair them completely. On the evidence of this week, that test is still open.
Sources
- Reuters — US Department of Agriculture breached by hackers this week, NBC News reports
- Interesting Engineering — FBI hack and USDA cyberattack target critical government data (reporting CNN’s account of the NCIC investigation)
- Seoul Economic Daily — FBI Probes Suspected Breach of National Crime Database
- FBI — NCIC Turns 50 (background on the National Crime Information Center)