Entities

Entities collected from this article, grouped for quick reference and copying. Inclusion does not establish malicious activity. Article links appear separately at the bottom as references.

Every day, the National Crime Information Center processes 10.8 million transactions. An officer enters a name or checks a vehicle or firearm. Within seconds, the system can return a wanted-person hit, a stolen-property record, or a protection order. It records the search too. The Federal Bureau of Investigation (FBI) is now investigating apparent unauthorized access and has not said whether the intruder reached criminal records, the transaction history, or one trusted route into either.

CNN reported on November 20, 2026 that the FBI was investigating an apparent breach of the National Crime Information Center (NCIC), citing two people briefed on the inquiry. Bureau officials had learned of the intrusion by Thursday, November 19. NBC News separately reported the investigation through two people briefed on the matter. The Associated Press independently confirmed the investigation on November 21 through a person familiar with it. The FBI declined to comment to the Associated Press, and the scope remained unclear.

NCIC connects criminal justice agencies across every US jurisdiction. Canada has direct access, while other foreign governments receive extracted information case by case through the US National Central Bureau of INTERPOL. Unauthorized activity by one participating agency account can reach NCIC without an intruder compromising the FBI's central infrastructure. Central access and stolen agency credentials create different victims and evidence. A compromised terminal, cloud identity, or insider creates another set of consequences.

President Trump called the reported intrusion "not good" on November 21, then speculated the breach could expose more on the "Russia Russia Russia hoax" and voter fraud. His comments supplied no evidence about what the intruder accessed. Politics began assigning meaning to the breach before the FBI had publicly established its location.

In the week before, a malfunction in the Federal Aviation Administration's network caused the longest nationwide ground stop in 25 years, and the National Security Agency suffered an outage. The Department of Agriculture took applications offline after a cyber security incident.

Ten Million Transactions a Day

The FBI's fiscal 2027 budget request describes NCIC as a 24-hour database containing more than 19.5 million active records across 22 files. Wanted and missing persons sit beside records on stolen property and protection orders. Other files cover gangs, sex offenders, immigration violators, and people under protective interest. The system averaged 99.8 percent uptime over the past 12 months, and police and courts rely on it during ordinary work. So do prosecutors, corrections officials, and investigators.

NCIC is unclassified, according to the FBI's May 2026 system-of-records notice. Unclassified does not mean public. The FBI's privacy assessment lists identifying information in the records ranging from names and photographs to passport details and Social Security numbers. They can also include medical characteristics, warrant information, and investigative identifiers. The system holds information about people named in active investigations by authorized users.

Participating agencies contribute the records and remain responsible for their accuracy and timeliness. Authorized users reach them through text-based queries built from names and descriptive data. This architecture distributes trust across a national user community. It also means the phrase "breach of NCIC" does not identify which security boundary failed.

Denmark supplies a close precedent. Threat actors did not need to compromise its central population register when they abused one trusted company's legitimate access to enumerate 8.8 million identities. NCIC is not Denmark's register, and no evidence links the incidents. The comparison shows why accepted credentials and an authorized connection can turn a central database into the apparent victim while the original compromise sits elsewhere.

CNN and NBC have not disclosed what their sources mean by breach. Neither report identifies an Originating Agency Identifier or user account. They name no terminal or application. No FBI system, cloud service, exploit, or affected record type appears either. The public evidence therefore supports an investigation into apparent unauthorized NCIC access. The stronger conclusion someone compromised the central environment remains unsupported.

The criminal records are only one collection inside NCIC. Its Transaction Log records every query and access. It also stores entries and updates, along with clearances and cancellations. Rejected transactions and administrative messages remain in the same history. The FBI now retains it for the life of the system, apart from history before 1990 that was kept for ten years.

A wanted-person record tells an intruder who is wanted. The transaction history can show who law enforcement looked for and when the search occurred. It can also identify the authorized route generating it. A foreign intelligence service can use those patterns to identify investigative interest before an arrest or public filing. An eCrime threat actor can warn fugitives or identify scrutiny around an associate. It can also sell access to someone deciding whether a name or vehicle is safe to use.

The same reversal followed a breach of Chainalysis, which exposed identity data surrounding the people who investigate cryptocurrency crime. The primary records were not required to create counter-intelligence value. The identities and relationships around the investigative system were enough to begin mapping its users. NCIC's transaction history can carry an even stronger signal because a query records investigative attention as it happens.

Administrative metadata can reveal a mission without exposing the mission file. The Icepick leak connected Department of Homeland Security personnel to sensitive programs through group names and reporting relationships. NCIC transactions create a different graph. Subjects connect to searches, which connect to authorized identities or agencies. Timing lets a service align those relationships with arrests and border encounters. A service can add events from its own collection.

The transaction history is also the FBI's strongest forensic advantage. Its privacy assessment says NCIC separately logs privileged user log-ins, role escalations, and failed password attempts. The system automatically generates alerts for unusual activity affecting personally identifiable information, and two designated security officers review system audit logs when alerts trigger or at least every seven calendar days. If those logs remain intact, the FBI should be able to distinguish one abused identity from central administrative access and bulk collection from a narrow query.

Three Access Paths Remain Open

Hypothesis 1, Authorized Agency Access

A threat actor compromised credentials, a terminal, or connected infrastructure belonging to an authorized agency and used the permission already attached to it. NCIC's distributed user base makes this the most economical access path in the absence of evidence showing a central exploit. A legitimate account would also explain why individual transactions could appear normal until their volume, timing, or targets formed an abnormal pattern.

The hypothesis cannot explain more than the account's permissions. NCIC access varies by agency and file, so one compromised identity may have reached only a narrow part of the system. The Originating Agency Identifier, terminal header, authentication history, and transaction volume would show whether the activity came through an established participant and how far its authority extended.

Hypothesis 2, Central NCIC or Cloud Access

The FBI's privacy assessment, approved in November 2022, said NCIC was transitioning to the Amazon Web Services government cloud. The May system-of-records notice says records may be kept in secure cloud computing environments, citing Federal Risk and Authorization Management Program (FedRAMP) approval as an example, and names Amazon Web Services as the provider on its publication date. The transition creates questions about cloud identities and administrative roles. Migration tooling, service configuration, and the boundary between FBI and provider responsibility require separate examination.

Nothing public implicates Amazon Web Services or establishes a cloud control-plane compromise. A central path would nevertheless carry the widest consequences because it can reach transaction history and privileged administration. It can also reach records beyond one agency's authorization. Cloud audit events and FBI administrative logs would help distinguish this hypothesis from one stolen participant account. Migration records and access spanning unrelated agencies would strengthen it.

Hypothesis 3, Insider Misuse

An authorized employee or contractor can query or extract data without defeating an external security control. NCIC's privacy assessment addresses the risk of unauthorized access and disclosure through role-escalation logging and access reviews. Automated alerts and regular audit reviews add more controls. A knowledgeable insider would understand which records matter and which searches attract attention. Ordinary workflows can then conceal collection.

Insider misuse usually leaves an identity trail unless the person borrows another account or tampers with logs. Schedule anomalies and unusual subjects would matter more than malware indicators. So would bulk queries, new privilege, and activity outside the user's assigned mission. The public reporting provides no evidence of an insider, so this remains a necessary alternative rather than a preferred explanation.

Stolen agency credentials can lead to privilege escalation. An insider can provide access later used from outside. A cloud identity can appear legitimate while reaching central resources. I assess with low confidence an authorized agency path is more likely than a novel compromise of NCIC itself. The judgment rests on the system's distributed trust model rather than evidence from this incident. The public record does not distinguish compromised credentials from insider misuse.

Motive Changes the Target

An eCrime threat actor would not need to copy 19.5 million records. Query access alone answers whether an associate is wanted or a vehicle has been reported stolen. It can also show whether law enforcement entered a person into a file. Persistent access can be sold to criminals who value one current answer more than a historical database dump. No public extortion demand, sale advertisement, or criminal claim had surfaced as of November 22.

A foreign intelligence service would have different collection requirements. Files covering known or suspected terrorists provide identifiers and status. So do records on foreign fugitives and immigration violators. Protective interests and active investigations add another layer. The transaction history then shows which subjects US agencies searched and when. A service seeking quiet access can monitor changes through repeated queries while avoiding the volume likely to trigger an alert.

The same service could also mine the records for people it can pressure. Outstanding arrest warrants, protection orders, supervised release, and gang records each mark legal exposure a handler can turn into leverage. The transaction history adds timing. A recent query shows whom law enforcement is checking now, possibly before the subject knows, the moment an offer of help or a threat of exposure lands hardest.

The privacy assessment also lists information on individuals subject to continuous evaluation or ongoing suitability determinations, which agencies collect for government benefits and employment background checks. Those entries cover the vetted population a service most wants, although only the agencies entering them can see them. Most NCIC subjects hold no value to a service. The value sits where NCIC hits overlap with personnel data, such as the FBI employee and applicant data the Associated Press reported ShinyHunters obtained in September.

The 2015 breach of the US Office of Personnel Management shows what such a match could start from. Threat actors took background investigation data on around 21.5 million people and 5.6 million sets of fingerprints. The stolen Standard Form 86 questionnaires hold the foreign contacts, family members, and financial history cleared personnel disclose. Matched against NCIC hits, data of that kind would point a service to a cleared person with a current legal problem.

An influence or hacktivist threat actor would gain value from exposure or manipulation. Publishing politically charged records can damage investigations or feed claims of selective enforcement. Altering a wanted-person record can create immediate fear, but the system places a control between a hit and official action. The NCIC operating manual requires agencies to confirm applicable hits with the agency entering the record. The check precedes an arrest, detention, property seizure, or several other listed actions.

Hit confirmation limits what one false entry can accomplish. The integrity risk survives it. A forged or canceled record can delay an officer, create a dangerous encounter, divert investigative effort, or become harder to detect if the originating agency is also compromised. The control works only when the second agency and its source documents remain trustworthy.

The reporting offers no basis to choose among eCrime, espionage, insider abuse, or politically motivated activity. No threat actor has claimed the incident. No government has been accused. The data types make several motives plausible, but target value cannot identify the person who reached it. Importing attribution from an earlier FBI breach would replace evidence with chronology.

Prior Breaches Supply No Attribution

ShinyHunters claimed the September compromise of the FBI's recruitment portal and theft of employee and applicant information. The FBI said the point of breach remained undetermined between a third party and its own enterprise. Director Kash Patel later said the compromise occurred on a platform managed by a third-party vendor and announced the arrest of an accused co-conspirator on October 9. None of the public reporting connects ShinyHunters to the new intrusion.

ShinyHunters operates as a contested eCrime identity rather than a stable organization with a reliable roster. People tied to the brand share handles and dispute membership. Reporting tied to the brand also includes allegations one participant planted another person's identity. Even a new ShinyHunters claim would require evidence linking the claimant to this access. The September incident alone supplies none.

CNN also recalled a 2023 compromise of a New York FBI field-office system used in child sexual exploitation investigations. Reuters later reported a foreign hacker reached files related to Jeffrey Epstein in that system, though its source described a cybercriminal rather than a foreign government. In March, the FBI also disclosed a probe of "suspicious activities" on an internal system holding sensitive surveillance and investigative information, the Associated Press recalled. Nothing in either incident identifies a common access path, threat actor, or objective.

The reporting itself has a narrower vantage point than many headlines imply. CNN and NBC cite people briefed on the investigation, while the Associated Press cites a person familiar with it. The additional source corroborates the inquiry but does not locate the access. None of the three outlets says its sources were technically positioned to distinguish central access from one participant connection. The phrase "FBI database breach" is therefore a reported description of the inquiry, not a published forensic finding.

What the FBI Can Prove

The FBI has published no indicators of compromise, so other agencies have nothing to block or search for. The evidence sits inside NCIC and agency audit records. Investigators can identify the Originating Agency Identifier and terminal header attached to each suspicious transaction.

The Criminal Justice Information Services (CJIS) Security Policy requires logging of successful and unsuccessful log-on attempts, password changes, and privileged-account actions. It lists query parameters and external credential use among the event types agencies consider, and criminal-history inquiries carry purpose codes. Investigators can compare those events with local agency logs and determine whether several unrelated agencies were accessed through one central identity. Without published values from any of those fields, participating agencies need behavioral scope before they can hunt reliably.

Containment should follow the access path rather than the headline. An exposed participant account requires credential revocation, terminal review, local log preservation, and a search for abnormal NCIC transactions under the same identity. Central administrative access requires a broader review of cloud roles, migration tooling, privileged sessions, and record changes. Suspected manipulation requires comparing entries, modifications, cancellations, and clearances with originating-agency source documents.

The FBI also needs to tell agencies whether the transaction history was queried or copied. A record exposure identifies subjects already inside NCIC. A transaction-log exposure identifies the law enforcement interest surrounding them. Those are different harms. The second can reveal investigations still developing, agencies working the same target, and changes in attention before an operation becomes public.

I assess with high confidence the public record establishes an FBI investigation into apparent unauthorized NCIC access. The public record does not support a compromise of the central NCIC environment because neither the FBI nor the reporting has located the access. It does not show whether anyone viewed or exfiltrated records. The same applies to alteration or deletion. The identity and its permissions would alter those judgments, as would the affected files. Transaction volume, access duration, and any record changes would provide the rest.

The FBI built NCIC to tell law enforcement what it knows about a person or case. Vehicles and firearms carry their own records. Its lifetime transaction history can tell an intruder something different. It can show what law enforcement wanted to know and when it asked. It can also show which part of government cared enough to search.

Until the FBI identifies the access path and accounts for that history, it cannot say whether this incident exposed the nation's crime records or the investigative attention surrounding them. If the transaction history left the system, a password reset can close the access. It cannot erase the map of US law enforcement interest already taken with it.

An agency connected to NCIC should revoke and reissue credentials for any account with abnormal NCIC transactions, review its terminals, and preserve local logs. The agency should also request a Transaction Log search from the FBI for queries run under its Originating Agency Identifier. Every entry, modification, cancellation, and clearance belongs against the originating agency's source documents, and every hit needs confirmation before an arrest or seizure. The FBI owes the central review and an answer on whether the transaction history was queried or copied.


Analyst Annex

Evidence status as of November 22, 2026. The records cover the reported Federal Bureau of Investigation (FBI) investigation into apparent unauthorized access to the National Crime Information Center (NCIC), the FBI and Department of Justice documentation governing NCIC architecture and auditing, and materially related FBI incidents kept separate below.

The FBI has not publicly confirmed the NCIC incident or identified the access path, affected identity, originating agency, accessed files, collection volume, record changes, threat actor, or indicators of compromise. ShinyHunters, the FBIJobs.gov compromise, the 2023 New York Field Office compromise, the March 2026 FBI internal-system incident, and Denmark's population-register exposure are context only.

Confirmed marks evidence published by a victim, vendor, regulator, or government authority about its own systems, decisions, or investigation. Reported marks technical findings or attribution assessments published by a third party. Claimed marks statements by a seller or self-identified threat actor that remain unverified. Provisional marks an analytical mapping supported by evidence but not made by the original source.

Indicators

The FBI has published no indicators of compromise, no Internet Protocol (IP) address, domain, account, exploit, vulnerability, filename, or file hash tied to the intruder. The table contains four NCIC transaction pivots from official documentation, the fields investigators would match against suspicious activity. They are standard schema values rather than evidence of malicious activity and must not be blocklisted. Each row points to the intruder only once investigators match it to the values in suspicious transactions.

IndicatorContextStatus
NCIC Transaction Audit Pivots
ORINine-character Originating Agency Identifier assigned by the FBI and used to control the information an agency can accessSource: FBI NCIC Privacy Impact AssessmentConfirmed
Transaction headerNine to nineteen characters containing routing information and retained as part of the NCIC Transaction LogSource: NCIC Operating ManualConfirmed
Purpose codeUsed with logging information to document why Interstate Identification Index information was soughtSource: FBI Criminal Justice Information Services Security PolicyConfirmed
QQQuery-query capability used to search previous NCIC inquiries involving a person, vehicle, license plate, or property during the preceding 30 daysSource: FBI NCIC Privacy Impact AssessmentConfirmed

MITRE ATT&CK Mapping

The public record establishes an investigation into apparent unauthorized access but discloses no access method, account use, query procedure, collection, persistence, exfiltration, or destructive action. No Enterprise ATT&CK technique can be mapped without converting the article's access hypotheses into observed behavior.

TacticTechniqueProcedureStatus
Reported NCIC Unauthorized Access
No technique mapped because the disclosed evidence contains no adversary procedure

Entities

EntityRoleStatus
Threat Actors
ShinyHunterseCrime threat actor, context onlyClaimed the separate September 2026 FBIJobs.gov compromise. No public source connects the name to the NCIC incidentSource: Associated Press, FBI, Associated Press on NCICClaimed
Government Bodies
Federal Bureau of InvestigationUS law enforcement agency and reported victimMaintains NCIC. Public reporting says it is investigating apparent unauthorized access, while the bureau declined to comment publiclySource: Associated PressReported
FBI Criminal Justice Information Services (CJIS) DivisionSystem owner and administratorOperates NCIC for federal, state, local, and tribal criminal justice agenciesSource: FBI NCIC Privacy Impact Assessment, FBI Fiscal 2027 Budget RequestConfirmed
NCIC Participating Criminal Justice AgenciesData contributors and authorized access pointsSubmit records, remain responsible for their accuracy, and reach NCIC through state or regional systems or direct connections. No participating agency has been identified as affectedSource: FBI NCIC data responsibilities, access architectureConfirmed
US National Central Bureau (USNCB) of INTERPOLInternational information-exchange routeReceives one direct NCIC connection and routes case-by-case extracts to foreign governments other than Canada. No public evidence identifies this path as affectedSource: FBI NCIC Privacy Impact AssessmentConfirmed
CanadaForeign government with direct NCIC accessThe only foreign government with direct access to NCIC. No public evidence identifies Canadian access or systems as affectedSource: FBI NCIC Privacy Impact AssessmentConfirmed
Organizations and Victims
Amazon Web Services (AWS)Cloud service providerNamed provider for the government cloud environment described in NCIC documentation. No public evidence implicates AWS or its control planeSource: Federal Register, FBI NCIC Privacy Impact AssessmentConfirmed
Tools and Services
National Crime Information CenterUnclassified criminal justice information systemCentral repository holding more than 19.5 million active records across 22 files and processing 10.8 million transactions each daySource: FBI Fiscal 2027 Budget RequestConfirmed
NCIC Transaction LogSystem audit and transaction historyRetains entries, updates, queries, access, clearances, cancellations, rejected transactions, and administrative messages for the life of the system, except pre-1990 historySource: Federal RegisterConfirmed
AWS Government Cloud EnvironmentDocumented NCIC cloud environmentThe 2022 privacy assessment says NCIC was transitioning to an AWS government cloud. The 2026 notice says records may be kept in secure cloud environmentsSource: FBI NCIC Privacy Impact Assessment, Federal RegisterConfirmed
Law Enforcement Enterprise PortalFederal law enforcement web portalHosts web interfaces authorized federal users can use to search NCIC. No public evidence identifies the portal as affectedSource: FBI NCIC Privacy Impact AssessmentConfirmed
NCIC Mobile ServiceRemote query interfaceAllows certified FBI users to request person, vehicle, article, and gun information but not enter, modify, cancel, or clear records. No public evidence identifies it as affectedSource: FBI NCIC Privacy Impact AssessmentConfirmed
Web Access Messaging System (WAMS)Internal FBI transaction interfaceAllows FBI divisions, task forces, and field offices to query NCIC and enter, modify, clear, or cancel records. No public evidence identifies it as affectedSource: FBI NCIC Privacy Impact AssessmentConfirmed
CJIS Shared Enterprise NetworkNetwork boundary and monitoring infrastructurePlaces firewalls, intrusion detection systems, and proxy devices between FBI and constituent systems and is monitored by CJIS and FBI security operations centersSource: FBI NCIC Privacy Impact AssessmentConfirmed
Interstate Identification Index (III)Criminal-history information serviceAdjacent service reachable through WAMS. No public evidence says the reported access reached itSource: FBI NCIC Privacy Impact AssessmentConfirmed
National Instant Criminal Background Check System (NICS)Firearms eligibility systemAdjacent system whose search criteria are excluded from the NCIC Transaction Log. No public evidence says NICS was affectedSource: Federal RegisterConfirmed
NletsInterstate public safety information networkConnected service available through FBI interfaces for state records including driver, vehicle, warrant, and criminal-history information. No public evidence says it was affectedSource: FBI NCIC Privacy Impact AssessmentConfirmed

The table covers separate FBI incidents and the Denmark population-register precedent. None of their threat actor claims, access paths, infrastructure, records, or technical artifacts was observed in the reported NCIC incident.

Intelligence Gaps

GapWhy It MattersWhat Would Close It
Compromised Security BoundaryDistinguishes central NCIC or cloud access from one participating agency, FBI interface, connected service, or insider accountFBI architecture statement, incident diagram, affected-system inventory, or forensic report naming the breached boundary
Originating Agency, Terminal, and Partner NotificationIdentifies whether suspicious transactions entered through an established participant, and participating agencies cannot scope local identities or preserve evidence without the affected transactions and time windowSuspicious ORI and transaction-header values matched to state, regional, federal, Canadian, or INTERPOL access records, plus FBI or CJIS partner notices and affected-agency disclosures
Account and Initial Access PathSeparates stolen credentials, an insider, a shared account, a compromised terminal, session theft, an exploited vulnerability, or a central administrative identity, and sets whether agencies prioritize credential revocation, endpoint response, vulnerability remediation, or insider investigationUser identifier, authentication method, device identity, source address, session and identity-provider logs, exploited vulnerability, malware, access-token history, or personnel finding
Accessed NCIC FilesSets the exposed population and distinguishes person files, property files, images, criminal histories, and transaction metadataMessage keys, file identifiers, query arguments, returned-record types, and affected-record counts
Transaction Log and Data ExfiltrationViewing, local export, and transfer create different harms and notification duties, and a copied Transaction Log would let the threat actor reconstruct law enforcement interest, investigative timing, and participating agenciesAudit records of Transaction Log queries, exports, or administrative access, plus egress telemetry, application programming interface (API) responses, download records, staging artifacts, and destination infrastructure
Access Duration and VolumeSeparates a narrow unauthorized query from persistent surveillance or bulk collection and changes the likely motiveFirst and last observed timestamps, query counts, result counts, rate patterns, and session duration
Record IntegrityAlteration, cancellation, or deletion can affect arrests, protection orders, missing-person cases, stolen-property recovery, and officer safetyEntry, update, clearance, and cancellation histories reconciled with originating-agency source documents
Cloud Environment InvolvementChanges the affected identities, telemetry owners, containment scope, and division of responsibility between the FBI and AWSAWS audit events, cloud identity records, administrative-role history, migration logs, and an FBI or AWS statement
Audit Log IntegrityEvery access-path judgment depends on transaction, authentication, privilege, and network records remaining complete and trustworthyLog-integrity validation, retention confirmation, clock reconciliation, missing-event analysis, and evidence of attempted deletion or modification
Query Notification and Replay ExposureDelayed inquiries, investigative-interest notifications, and the QQ capability can reveal or replay recent law enforcement interest and may also expose abnormal queryingMessage-key history, investigative-interest flags, delayed-inquiry matches, QQ searches, and related notifications
Connected System ReachFBI interfaces and routing headers distinguish NCIC, III, NICS, and Nlets, but no public source says the reported activity remained inside NCICRouting prefixes such as 2L01 for NGI and Interstate Identification Index traffic and 6L01 for NICS, interface logs, cross-system query records, and scope statements for III, NICS, Nlets, and NGI
Foreign Access ExposureDirect Canadian access and INTERPOL exchanges expand the possible victim and intelligence-sharing consequences beyond US agenciesCanadian and USNCB connection logs, partner notifications, extracted-record histories, and government scope statements
Threat Actor Identity and MotiveeCrime query resale, counter-intelligence collection, influence activity, and insider misuse imply different collection patterns and response prioritiesAttributed infrastructure, account ownership, communications, payment records, victim selection, collection behavior, or law enforcement findings