preventions
- ID: PV016
- Created: 01st June 2024
- Updated: 19th July 2024
- Contributor: The ITM Team
Enforce a Data Classification Policy
A Data Classification Policy establishes a standard for handling data by setting out criteria for how data should be classified and subsequently managed and secured. A classification can be applied to data in such a way that the classification is recorded in the body of the data (such as a footer in a text document) and/or within the metadata of a file.
Sections
| ID | Name | Description |
|---|---|---|
| IF002 | Exfiltration via Physical Medium | A subject exfiltrates organizational information through the unauthorized transfer or removal of physical storage media, computing devices, or documents containing organizational information. This may involve copying data to removable storage devices, transferring information between directly connected systems, accessing internal storage drives, or physically removing documents from the organization's control.
Physical exfiltration methods may include USB mass storage devices, optical or floppy disk media, additional internal drives, personal mobile devices, and direct access to a target system's storage. The subject may also remove physical documents containing sensitive or confidential information without converting them into a digital format. |
| IF003 | Exfiltration via Media Capture | Exfiltration via media capture refers to the extraction of sensitive information through the recording of visual or auditory content using capture mechanisms that operate outside organizational control. This includes the use of external devices, embedded system tools, or third-party applications to record screens, documents, or conversations and convert them into transferable media formats such as images, video, audio, or structured transcripts.
This category is defined not by the type of data being accessed, but by the method of extraction, specifically, the transformation of information into captured media in order to bypass conventional monitoring and control mechanisms. In these scenarios, the subject does not transfer files or data through approved or monitored channels. Instead, they reproduce the information in an alternate form that can be removed without generating traditional indicators of exfiltration.
Media capture techniques are particularly effective in environments where digital controls are mature, such as strong data loss prevention (DLP), restricted file transfer mechanisms, or monitored endpoints. As these controls limit conventional exfiltration paths, subjects may shift toward out-of-band capture methods that operate beyond system visibility.
This behavior may be opportunistic or deliberate. In lower-control environments, subjects may casually capture information with minimal consideration of detection. In higher-control environments, the use of media capture may indicate awareness of monitoring capabilities and an intentional effort to circumvent them. In both cases, the technique exploits a fundamental gap between information exposure and information control, once data is visible or spoken, it becomes inherently difficult to contain.
Media capture also varies in its execution and detectability. Some techniques are rapid and discrete, such as still photography, while others involve sustained collection, such as video recording or continuous audio capture.
From an investigative perspective, this section represents a class of behaviors where traditional telemetry is limited or absent. Detection often relies on indirect indicators, environmental controls, or post-event analysis of leaked material. As a result, prevention and deterrence play a critical role, particularly through physical controls, policy enforcement, and attribution mechanisms such as watermarking. This section is closely related to broader data loss behaviors, but is distinct in its reliance on out-of-band capture methods rather than direct data transfer . |
| IF004 | Exfiltration via Other Network Medium | A subject exfiltrates files through a network. A network can be an Internet Protocol (IP) network or other technology enabling the communication of data between two or more digital devices. |
| ME014 | Printing | A subject has the ability to print organizational documents, images, or other information using a printer accessible from an organizational device. This capability enables digital information to be reproduced in a physical format, including documents containing confidential business information, intellectual property, financial records, or other sensitive organizational material.
Printing may occur through organization-managed printers or external printing devices, depending on the subject's access and available printing configurations. The production of physical copies may enable information to be retained, reproduced, or transported independently of the original digital system and its associated access controls. |
| MT008 | Lack of Awareness | A subject undertakes actions that constitute an infringement because they are unaware of applicable organizational policies, security requirements, or restrictions governing their conduct. The subject may believe their actions are permitted or fail to recognize that particular activities, such as transferring organizational information to personal accounts, using unapproved services, or accessing restricted resources, are prohibited.
Lack of awareness may arise from insufficient communication of policy requirements, unclear organizational guidance, or a misunderstanding of the restrictions applicable to the subject's role. Unlike recklessness, the subject does not knowingly disregard an understood risk or restriction, but acts without sufficient knowledge that their conduct is prohibited.
During an investigation, the subject's awareness of relevant policies and restrictions should be assessed through available evidence, including policy acknowledgments, previous communications, guidance provided, and prior enforcement actions. An infringement should not automatically be attributed to lack of awareness solely because the subject claims they did not know the activity was prohibited. |
| IF018 | Sharing on AI Chatbot Platforms | A subject discloses sensitive organizational information through interactions with external Artificial Intelligence (AI) chatbot platforms. This may involve submitting confidential documents, source code, customer information, internal communications, or other restricted information through text prompts, file uploads, or additional input capabilities provided by the platform.
The disclosure may be intentional, where the subject deliberately transfers organizational information to an external AI service, or reckless, where sensitive information is submitted without adequate consideration of organizational restrictions or the platform's data handling arrangements. Depending on the service configuration, submitted information may be processed, retained, or made accessible within systems outside the organization's authorized information governance environment.
Sharing sensitive information through AI chatbot platforms may result in unauthorized data disclosure, loss of confidentiality, or violations of organizational policies and contractual obligations. The investigative significance depends on the information submitted, the subject's authorization, the service's data handling arrangements, and whether the disclosure was intentional or reckless. |
| ME023 | Sensitivity Label Leakage | Sensitivity label leakage refers to the exposure or misuse of classification metadata, such as Microsoft Purview Information Protection (MIP) sensitivity labels, through which information about the nature, importance, or confidentiality of a file is unintentionally or deliberately disclosed. While the underlying content of the document may remain encrypted or otherwise protected, the presence and visibility of sensitivity labels alone can reveal valuable contextual information to an insider.
This form of leakage typically occurs when files labeled with sensitivity metadata are transferred to insecure locations, shared with unauthorized parties, or surfaced in logs, file properties, or collaboration tool interfaces. Labels may also be leaked through misconfigured APIs, email headers, or third-party integrations that inadvertently expose metadata fields. The leakage of sensitivity labels can help a malicious insider identify and prioritize high-value targets or navigate internal systems with greater precision, without needing immediate access to the protected content.
Examples of Use:
|
| MT016 | Human Error | The subject has no threatening motive and is not reckless in their actions. The infringement is a result of an honest mistake made by the subject. |
| MT013 | Misapprehension or Delusion | A subject is motivated to undertake an infringement as a result of a materially incorrect understanding of circumstances, events, or perceived threats. The subject's actions are based on factual assumptions that are demonstrably false or inconsistent with established evidence, rather than an accurate understanding of the situation.
This may include a subject who incorrectly believes organizational systems have been compromised, that colleagues are acting against them, or that particular information or assets present a threat requiring intervention. Acting on these misconceptions, the subject may undertake actions they believe are necessary or justified, despite the absence of the circumstances they perceive.
During an investigation, the subject's understanding of events should be assessed against corroborating evidence. Investigators should establish whether a material factual misconception influenced the infringement and avoid inferring a psychiatric condition solely from the subject's statements or behavior. |
| IF028 | Delegated Execution via Artificial Intelligence Agents | A subject causes organizational harm by delegating the execution of an infringement to an artificial intelligence (AI) agent, including in circumstances where the agent possesses access or authority beyond the subject’s own direct permissions.
This behavior occurs when a subject authorizes, configures, or directs an AI agent to perform operational actions inside the trusted environment that result in unauthorized access, data loss, fraud, operational disruption, or other policy-violating impact. The AI agent executes the harmful activity on the subject’s behalf.
An AI agent, in this context, is a system capable of autonomously or semi-autonomously performing structured tasks, interacting with enterprise systems, invoking APIs, chaining actions across platforms, or maintaining persistent monitoring logic. Unlike simple prompt-based AI use, the agent is empowered to act within organizational systems.
In certain environments, AI agents are deployed with elevated or system-level permissions to enable productivity, indexing, analytics, or workflow automation. A subject may intentionally leverage this broader authority to access data, systems, or functionality that exceeds their own interactive role-based access. When such delegated activity results in material harm, it constitutes an infringement under this Section.
Examples include:
The defining characteristic is that the harmful act is executed by the AI agent under authority granted or exploited by the subject. The subject extends the organization’s trust boundary to an autonomous system and operationalizes it to inflict harm.
The subject remains fully accountable for the resulting impact. The AI agent amplifies speed, scale, and efficiency, and in cases of elevated agent permissions, may enable privilege amplification beyond the subject’s direct access. |
| IF045 | Unauthorized Data Access | A subject accesses, views, searches, queries, retrieves, or otherwise examines organizational data without a legitimate business purpose or authorization to do so. The subject may use their own valid account and may possess the technical permissions required to access the data; however, the access falls outside their approved role, assigned duties, operational need, or other authorized purpose.
Unauthorized Data Access may involve customer records, employee information, commercially sensitive material, security information, investigative records, financial data, intellectual property, or other restricted organizational information. Examples include accessing the records of colleagues, acquaintances, public figures, or customers without a business requirement; searching sensitive datasets out of curiosity or personal interest; or reviewing information for an external, financial, or otherwise unauthorized purpose.
The infringement occurs through the unauthorized access itself and does not require the subject to modify, disclose, copy, or exfiltrate the information. Where the subject subsequently transfers, exposes, alters, or otherwise misuses the accessed data, additional Infringement Sections or Sub-sections may also apply. |
| PR054 | Record Export | A subject uses reporting, query, or export functions to generate a separately usable copy of records held within an organizational application or database in preparation for an infringement. The subject may run an existing report, select particular fields or record populations, expand reporting parameters, or perform repeated exports to assemble the required information. The records may concern customers, personnel, financial transactions, cases, or other organizational activity. The resulting extract may be retained as a report, spreadsheet, data file, or saved dataset, including within an organizational system without being downloaded to the subject’s device.
The defining behavior is the production of a record-level extract from underlying application or database content, rather than the retrieval of pre-existing files. The subject may use legitimately assigned permissions without circumventing a technical control. Investigators should establish which records and fields were included, whether the output was successfully generated and retained, and how its creation relates to the subject’s responsibilities, approved tasks, and subsequent activity. Possession of export permissions, creation of a report definition, or ordinary viewing of records does not independently establish this behavior. An export’s size or sensitivity alone does not establish preparation for an infringement. |
| IF001.004 | Exfiltration via Webhook | A subject exfiltrates organizational data by transmitting it to an external webhook endpoint, typically through HTTP POST requests. Webhooks allow applications and services to receive data automatically when triggered by an event or invoked directly through a script, application, or command-line tool.
The subject may configure or misuse webhook integrations associated with services such as Slack, Discord, or Microsoft Teams to transfer sensitive information outside authorized organizational boundaries. Data may be transmitted within structured request payloads, enabling automated or repeated exfiltration without conventional file uploads. |
| IF001.001 | Exfiltration via Cloud Storage | A subject uses a cloud storage service, such as Dropbox, OneDrive, or Google Drive to exfiltrate data. They will then access that service again on another device to retrieve the data. Examples include (URLs have been sanitized):
|
| IF001.002 | Exfiltration via Code Repository | A subject uses a code repository service, such as GitHub, to exfiltrate data. They will then access that service again on another device to retrieve the data. Examples include (URLs have been sanitized):
|
| IF001.003 | Exfiltration via Text Storage Sites | A subject uses a text storage service, such as Pastebin, to exfiltrate data. They will then access that service again on another device to retrieve the data. Examples include (URLs have been sanitized):
|
| IF002.001 | Exfiltration via USB Mass Storage Device | A subject exfiltrates data using a USB-connected mass storage device, such as a USB flash drive or USB external hard-drive. |
| IF002.002 | Exfiltration via Physical Access to System Drive | A subject exfiltrates organizational information by physically accessing the internal storage drive of a target system and transferring its contents outside authorized organizational control. This may involve removing a hard disk drive (HDD) or solid-state drive (SSD) from a workstation, server, or other device and connecting it to another computer to access or copy stored information.
Direct access to the storage device may allow the subject to examine or extract information independently of the original operating system and its associated monitoring controls. The extent of accessible information depends on the storage configuration, file system, encryption protections, and availability of the credentials required to decrypt protected data. |
| IF002.003 | Exfiltration via New Internal Drive | A subject exfiltrates data by connecting an additional drive to a system using the Serial Advanced Technology Attachment (SATA) interface on a motherboard, and copying files to the new storage device. |
| IF002.004 | Exfiltration via Floppy Disk | A subject exfiltrates data using a floppy disk drive. |
| IF003.001 | Exfiltration via Photography | A subject captures sensitive information by taking still images using an external device, most commonly a personal mobile phone. This typically involves photographing screens, printed documents, whiteboards, or other visual representations of sensitive data within the organization’s environment.
Unlike video capture, photography enables rapid, low-friction extraction of discrete information with minimal dwell time. A subject can capture high volumes of content in short bursts without sustained or conspicuous behavior, making this technique particularly effective in environments with physical proximity to sensitive material but strong digital controls.
This method often operates entirely outside controlled systems and therefore bypasses endpoint monitoring, data loss prevention (DLP), and network-based detection mechanisms. It is frequently opportunistic, occurring during routine access to sensitive information, but may also be deliberate, such as systematically photographing documents, screens, or workflows over time.
Photography-based exfiltration is especially prevalent in environments where:
The presence of this behavior may indicate awareness of monitoring controls or a preference for low-risk, low-detectability exfiltration methods. |
| IF003.002 | Exfiltration via External Device Video Capture | A subject records sensitive information by capturing video using an external device, such as a personal mobile phone or standalone camera. This behavior typically involves filming screens, documents, or physical environments where sensitive information is displayed or discussed.
Unlike software-based screen recording or screenshot tools, this method operates outside corporate control boundaries. The capture process occurs entirely outside the monitored endpoint, bypassing data loss prevention (DLP), endpoint detection, and audit logging mechanisms.
This technique is commonly observed in controlled environments where digital exfiltration is restricted or heavily monitored. It may be opportunistic (such as quickly recording a screen) or deliberate, involving repeated capture of large volumes of information over time. The use of an external device can indicate subject awareness of monitoring controls and an intent to avoid traceable data transfer. |
| IF003.003 | Exfiltration via Audio Capture | A subject captures sensitive information by recording audio using an external device, most commonly a personal mobile phone or wearable device. This typically involves recording conversations, meetings, phone calls, or ambient discussions where sensitive information is disclosed verbally.
Unlike visual capture techniques, audio capture does not require direct interaction with systems or documents. It enables the subject to collect information passively, often without needing to position a device toward a specific target. As a result, this method can be sustained over longer periods with reduced risk of detection, particularly in collaborative or discussion-heavy environments.
This technique operates entirely outside corporate monitoring controls, bypassing endpoint telemetry, data loss prevention (DLP), and access logging. It is particularly effective in environments where sensitive information is frequently communicated verbally, including meetings, support operations, incident response discussions, executive briefings, and informal conversations between colleagues.
Audio capture is often deliberate, as it requires forethought to record and later process the information. However, it may also be opportunistic, especially where subjects are routinely exposed to sensitive discussions. The presence of this behavior may indicate an intent to capture information that is not otherwise accessible in written or exportable form. |
| IF004.001 | Exfiltration via Bluetooth | A subject exfiltrates files using Bluetooth as the transportation medium. |
| IF004.002 | Exfiltration via AirDrop | A subject exfiltrates files using AirDrop as the transportation medium. |
| IF005.001 | Exfiltration via Installed Messaging Application | A subject exfiltrates organizational information using a messaging application installed on a device. The information may be transmitted to another account or individual, or retained within the messaging service for subsequent access from another device. |
| IF005.002 | Exfiltration via Web-Based Messaging Application | A subject exfiltrates organizational information using a web-based messaging application accessed through a browser. The information may be transmitted to another account or individual, or retained within the messaging service for subsequent access from another device. |
| IF002.005 | Exfiltration via Physical Documents | A subject transports physical documents outside of the control of the organization. |
| ME014.001 | External Printing | A subject has the ability to print organizational documents or other information using a printer that is not owned, managed, or controlled by the organization. This may include personal printers, home printing equipment, or third-party printing services accessible through local connections, wireless networks, or remote printing functionality.
External printing enables the subject to produce physical copies of organizational information outside established printing infrastructure and associated monitoring or access restrictions. Depending on the printer and service used, printed materials may be retained by the subject or become accessible to individuals outside the organization. |
| IF004.003 | Exfiltration via Personal NAS Device | A subject exfiltrates data using an organization-owned device (such as a laptop) by copying the data from the device to a personal Network Attached Storage (NAS) device, which is attached to a network outside of the control of the organization, such as a home network. Later, using a personal device, the subject accesses the NAS to retrieve the exfiltrated data. |
| IF002.006 | Exfiltration via USB to USB Data Transfer | A USB to USB data transfer cable is a device designed to connect two computers directly together for the purpose of transferring files between them. These cables are equipped with a small electronic circuit to facilitate data transfer without the need for an intermediate storage device. Typically a USB to USB data transfer cable will require specific software to be installed to facilitate the data transfer. In the context of an insider threat, a USB to USB data transfer cable can be a tool for exfiltrating sensitive data from an organization's environment. |
| IF001.005 | Exfiltration via Note-Taking Web Services | A subject uploads confidential organization data to a note-taking web service, such as Evernote. The subject can then access the confidential data outside of the organization from another device. Examples include (URLs have been sanitized):
|
| IF004.004 | Exfiltration via Screen Sharing Software | A subject exfiltrates data outside of the organization's control using the built-in file transfer capabilities of software such as Teamviewer. |
| IF018.001 | Exfiltration via AI Chatbot Platform History | A subject exfiltrates sensitive organizational information by submitting it to an external Artificial Intelligence (AI) chatbot platform that retains conversation history. The information may remain accessible through the subject's account, allowing subsequent retrieval from another device or location. Submission of the information to the external platform constitutes the exfiltration, regardless of whether it is subsequently retrieved. |
| IF018.002 | Reckless Sharing on AI Chatbot Platforms | A subject recklessly interacts with a public Artificial Intelligence (AI) chatbot (such as ChatGPT and xAI Grok), leading to the inadvertent sharing of sensitive information. The submission of sensitive information to public AI platforms risks exposure due to potential inadequate data handling or security practices. Although some platforms are designed not to retain specific personal data, the reckless disclosure could expose the information to unauthorized access and potential misuse, violating data privacy regulations and leading to a loss of competitive advantage through the exposure of proprietary information. |
| IF002.010 | Exfiltration via Bring Your Own Device (BYOD) | A subject connects their personal device, under a Bring Your Own Device (BYOD) policy, to organization resources, such as on-premises systems or cloud-based platforms. By leveraging this access, the subject exfiltrates sensitive or confidential data. This unauthorized data transfer can occur through various means, including copying files to the personal device, sending data via email, or using cloud storage services. |
| IF022.002 | PII Leakage (Personally Identifiable Information) | PII (Personally Identifiable Information) leakage refers to the unauthorized disclosure, exposure, or mishandling of information that can be used to identify an individual, such as names, addresses, phone numbers, national identification numbers, financial data, or biometric records. In the context of insider threat, PII leakage may occur through negligence, misconfiguration, policy violations, or malicious intent.
Insiders may leak PII by sending unencrypted spreadsheets via email, exporting user records from customer databases, misusing access to HR systems, or storing sensitive personal data in unsecured locations (e.g., shared drives or cloud storage without proper access controls). In some cases, PII may be leaked unintentionally through logs, collaboration platforms, or default settings that fail to mask sensitive fields.
The consequences of PII leakage can be severe—impacting individuals through identity theft or financial fraud, and exposing organizations to legal penalties, reputational harm, and regulatory sanctions under frameworks such as GDPR, CCPA, or HIPAA.
Examples of Infringement:
|
| IF022.003 | PHI Leakage (Protected Health Information) | PHI Leakage refers to the unauthorized, accidental, or malicious exposure, disclosure, or loss of Protected Health Information (PHI) by a healthcare provider, health plan, healthcare clearinghouse (collectively, "covered entities"), or their business associates. Under the Health Insurance Portability and Accountability Act (HIPAA) in the United States, PHI is defined as any information that pertains to an individual’s physical or mental health, healthcare services, or payment for those services that can be used to identify the individual. This includes medical records, treatment history, diagnosis, test results, and payment details.
HIPAA imposes strict regulations on how PHI must be handled, stored, and transmitted to ensure that individuals' health information remains confidential and secure. The Privacy Rule within HIPAA outlines standards for the protection of PHI, while the Security Rule mandates safeguards for electronic PHI (ePHI), including access controls, encryption, and audit controls. Any unauthorized access, improper sharing, or accidental exposure of PHI constitutes a breach under HIPAA, which can result in significant civil and criminal penalties, depending on the severity and nature of the violation.
In addition to HIPAA, other countries have established similar protections for PHI. For example, the General Data Protection Regulation (GDPR) in the European Union protects personal health data as part of its broader data protection laws. Similarly, Canada's Personal Information Protection and Electronic Documents Act (PIPEDA) governs the collection, use, and disclosure of personal health information by private-sector organizations. Australia also has regulations under the Privacy Act 1988 and the Health Records Act 2001, which enforce stringent rules for the handling of health-related personal data.
This infringement occurs when an insider—whether maliciously or through negligence—exposes PHI in violation of privacy laws, organizational policies, or security protocols. Such breaches can involve unauthorized access to health records, improper sharing of medical information, or accidental exposure of sensitive health data. These breaches may result in severe legal, financial, and reputational consequences for the healthcare organization, including penalties, lawsuits, and loss of trust.
Examples of Infringement:
|
| IF023.002 | Sanction Violations | Sanction violations involve the direct or indirect engagement in transactions with individuals, entities, or jurisdictions that are subject to government-imposed sanctions. These restrictions are typically enforced by regulatory bodies such as the U.S. Department of the Treasury’s Office of Foreign Assets Control (OFAC), the United Nations, the European Union, and equivalent authorities in other jurisdictions.
Unlike export violations, which focus on the control of goods and technical data, sanction violations concern the status of the receiving party. A breach occurs when a subject facilitates, authorizes, or executes transactions that provide economic or material support to a sanctioned target—this includes sending payments, delivering services, providing access to infrastructure, or sharing non-controlled information with a restricted party.
Insiders may contribute to sanction violations by bypassing compliance checks, falsifying documentation, failing to screen third-party recipients, or deliberately concealing the sanctioned status of a partner or entity. Such conduct can occur knowingly or as a result of negligence, but in either case, it exposes the organization to serious legal and financial consequences.
Regulatory enforcement for sanctions breaches may result in significant penalties, asset freezes, criminal prosecution, and reputational damage. Organizations are required to maintain robust compliance programs to monitor and prevent insider-driven violations of international sanctions regimes. |
| IF023.003 | Anti-Trust or Anti-Competition | Anti-trust or anti-competition violations occur when a subject engages in practices that unfairly restrict or distort market competition, violating laws designed to protect free market competition. These violations can involve a range of prohibited actions, such as price-fixing, market division, bid-rigging, or the abuse of dominant market position. Such behavior typically aims to reduce competition, manipulate pricing, or create unfair advantages for certain businesses or individuals.
Anti-competition violations may involve insiders leveraging their position to engage in anti-competitive practices, often for personal or corporate gain. These violations can result in significant legal and financial penalties, including fines and sanctions, as well as severe reputational damage to the organization involved.
Examples of Anti-Trust or Anti-Competition Violations:
Regulatory Framework:
Anti-trust or anti-competition laws are enforced globally by various regulatory bodies. In the United States, the Federal Trade Commission (FTC) and the Department of Justice (DOJ) regulate anti-competitive behavior under the Sherman Act, the Clayton Act, and the Federal Trade Commission Act. In the European Union, the European Commission enforces anti-trust laws under the Treaty on the Functioning of the European Union (TFEU) and the Competition Act. |
| IF022.004 | Payment Card Data Leakage | A subject with access to payment environments or transactional data may deliberately or inadvertently leak sensitive payment card information. Payment Card Data Leakage refers to the unauthorized exposure, transmission, or exfiltration of data governed by the Payment Card Industry Data Security Standard (PCI DSS). This includes both Cardholder Data (CHD)—such as the Primary Account Number (PAN), cardholder name, expiration date, and service code—and Sensitive Authentication Data (SAD), which encompasses full track data, card verification values (e.g., CVV2, CVC2, CID), and PIN-related information.
Subjects with privileged, technical, or unsupervised access to point-of-sale systems, payment gateways, backend databases, or log repositories may mishandle or deliberately exfiltrate CHD or SAD. In some scenarios, insiders may exploit access to system-level data stores, intercept transactional payloads, or scrape logs that improperly store SAD in violation of PCI DSS mandates. This may include exporting payment data in plaintext, capturing full card data from logs, or replicating data to unmonitored environments for later retrieval.
Weak controls, such as the absence of data encryption, improper tokenization of PANs, misconfigured retention policies, or lack of field-level access restrictions, can facilitate misuse by insiders. In some cases, access may be shared or escalated informally, bypassing formal entitlement reviews or just-in-time provisioning protocols. These gaps in security can be manipulated by a subject seeking to leak or profit from payment card data.
Insiders may also use legitimate business tools—such as reporting platforms or data exports—to intentionally bypass obfuscation mechanisms or deliver raw payment data to unauthorized recipients. Additionally, compromised service accounts or insider-created backdoors can provide long-term persistence for continued exfiltration of sensitive data.
Data loss involving CHD or SAD often trigger mandatory breach disclosures, regulatory scrutiny, and severe financial penalties. They also pose reputational risks, particularly when data loss undermines consumer trust or payment processing agreements. In high-volume environments, even small-scale leaks can result in widespread exposure of customer data and fraud. |
| IF022.001 | Intellectual Property Theft | A subject misappropriates, discloses, or exploits proprietary information, trade secrets, creative works, or internally developed knowledge obtained through their role within the organization. This form of data loss typically involves the unauthorized transfer or use of intellectual assets—such as source code, engineering designs, research data, algorithms, product roadmaps, marketing strategies, or proprietary business processes—without the organization's consent.
Intellectual property theft can occur during employment or around the time of offboarding, and may involve methods such as unauthorized file transfers, use of personal storage devices, cloud synchronization, or improper sharing with third parties. The consequences can include competitive disadvantage, breach of contractual obligations, and significant legal and reputational harm. |
| IF028.001 | AI Agent Internal Reconnaissance | A subject causes harm by directing an artificial intelligence (AI) agent to generate unauthorized internal intelligence about the organization—information the subject would not reasonably possess through legitimate role-based access or business need.
This behavior occurs when an AI agent is used to systematically query, correlate, or infer sensitive internal information by leveraging its broad integrations, indexing authority, or analytical capabilities. The resulting intelligence may reveal restricted projects, executive decisions, legal matters, acquisition activity, investigation status, architectural weaknesses, or other sensitive organizational insight.
Unlike routine search or manual browsing, AI agent internal reconnaissance enables:
In certain enterprise deployments, authorized AI platforms possess integration-level access across knowledge bases, ticketing systems, document repositories, messaging platforms, and identity directories. A subject may deliberately leverage this aggregated visibility to extract or infer intelligence beyond their legitimate business scope.
Examples include:
The infringement is established when the agent-generated intelligence materially exceeds legitimate business need and results in unauthorized exposure of sensitive organizational insight. The harm lies in the unauthorized acquisition of internal intelligence—particularly where that intelligence enables subsequent exploitation, trading, coercion, or strategic misuse.
The defining characteristic is not merely access, but computational synthesis. The AI agent transforms distributed internal data into actionable intelligence that the subject could not reasonably derive manually within policy boundaries. |
| PR035.001 | AI Agent Data Staging | A subject prepares for potential insider activity by directing an artificial intelligence (AI) agent to aggregate, organize, or transform sensitive organizational data into structured or portable formats.
This behavior occurs when an AI agent is tasked with systematically collecting information from internal repositories and consolidating it into outputs that are easier to store, review, transfer, or exploit. The agent performs bulk summarization, data normalization, or cross-repository aggregation that significantly reduces the effort required to later misuse the information.
Unlike reconnaissance activities that focus on discovering intelligence, AI Agent Data Staging focuses on operational preparation of data. The AI agent converts dispersed or complex internal information into consolidated outputs that increase its portability, usability, or accessibility outside its original context.
Examples include:
The defining characteristic of this Sub-section is the delegated consolidation of sensitive information. The subject leverages the AI agent to perform scalable data preparation that increases the volume, portability, or usability of organizational data.
While the staged data may not yet have been transferred outside the organization, the consolidation process materially lowers the effort required to exfiltrate or exploit it. In environments where AI platforms possess broad repository visibility, this capability can significantly accelerate the preparation phase of insider activity. |
| AF008.002 | Audio Steganography | A subject embeds data within an audio file or audio stream so that the concealed information is carried inside apparently legitimate sound content. The audio may continue to play normally, while modifications to the underlying samples, frequency components, timing, phase, or file structure contain a hidden payload.
Audio steganography may be used to conceal sensitive organizational data within music, voice recordings, podcasts, meeting recordings, sound effects, or other audio content before the file is transferred through an approved or commonly used communication channel. The subject may select audio because normal variations in sound, background noise, and compression can make small alterations difficult to identify through ordinary review.
Methods may include modifying the least significant bits of audio samples, altering phase or frequency characteristics, inserting information into low-audibility ranges, using echo-based encoding, or appending data within fields not required for normal playback. The concealed payload may also be encrypted before embedding, preventing recovery of its contents even where investigators identify that steganography has been used. |
| AF008.003 | Video Steganography | A subject embeds data within a video file or video stream to conceal sensitive information inside apparently legitimate visual or audiovisual content.
Video files provide multiple locations for hidden data, including individual frames, color values, compression coefficients, motion data, audio tracks, subtitles, metadata, and unused container structures. A subject may distribute a payload across many frames or components, reducing the amount of alteration applied to any single part of the file.
The resulting video may appear and play normally while carrying documents, source code, credentials, records, archives, or other organizational data. The subject may transfer the video through email, cloud storage, messaging platforms, or media-sharing services where large audiovisual files are expected. |
| AF008.004 | Text Steganography | A subject conceals information within apparently ordinary written text by manipulating characters, spacing, formatting, language patterns, or other textual features that are unlikely to attract attention during normal reading.
The visible content may appear to be a legitimate email, document, chat message, report, source-code comment, or other written communication. Hidden information may be encoded through additional spaces or tabs, capitalization, punctuation, selected words, spelling variations, character placement, font changes, Unicode substitutions, zero-width characters, or bidirectional control characters.
Text steganography may be used to communicate small quantities of sensitive information, instructions, identifiers, authentication material, or encoded content without creating an obvious attachment or visible block of ciphertext. |
| AF008.005 | Document Steganography | A subject conceals data within a business document by using content, objects, fields, formatting, or internal structures that are not visible during ordinary viewing.
The carrier may be a word-processing document, spreadsheet, presentation, Portable Document Format file, or another document type routinely used by the organization. The document may open and display expected content while containing hidden text, files, archives, scripts, worksheets, comments, embedded objects, metadata, or other concealed material.
Methods may include white text on a white background, extremely small font sizes, content positioned outside the visible page, hidden spreadsheet rows or worksheets, concealed presentation objects, embedded files, unused form fields, custom metadata, document layers, or internal package relationships. |
| AF008.006 | File-Structure Steganography | A subject conceals data within unused, ignored, optional, or non-rendered areas of a file or container so that the carrier continues to function normally while the hidden payload remains outside the content presented to the viewer.
The technique may use padding, reserved fields, unused headers, slack areas, archive comments, trailing data, application-specific metadata, or other structures retained during transfer but not normally displayed or processed. Examples include placing data after the logical end of a file, storing content in unused container fields, or exploiting differences between how applications and security tools interpret the same format.
This behavior may frustrate investigation because the visible content appears benign and the concealed data may only be identified through structural parsing, hexadecimal examination, file carving, or validation against the relevant file-format specification. |
| PR040.001 | Testing Data Transfer Controls | A subject conducts a limited transfer of non-sensitive or low-value data to determine whether organizational controls detect, block, quarantine, or escalate the activity.
The subject may send a file to a personal email account, upload content to an unapproved cloud storage service, transfer data through a messaging platform, or use another external destination before attempting a larger or more sensitive transfer. They may vary the file type, size, classification, destination, compression, encryption, or transfer method to identify which conditions trigger Data Loss Prevention (DLP), web proxy, email gateway, or cloud access security controls.
The behavior may include a pause after the test while the subject waits to determine whether security personnel, management, or another authority responds. A successful test may establish a viable exfiltration route or reveal thresholds that can be avoided during a later infringement. |
| IF045.001 | Unauthorized Access to Customer Data | A subject accesses, views, searches, queries, or retrieves customer data without a legitimate business purpose or authorization to do so. The subject may possess valid technical access to the relevant system or dataset but uses that access outside the requirements of their role, assigned duties, customer relationships, cases, transactions, or other approved activity.
Customer data may include personally identifiable information, account details, transactional records, financial information, support interactions, communications, location information, or other information relating to customers, clients, patients, subscribers, or account holders. The infringement occurs through the unauthorized access itself and does not require the information to be subsequently disclosed, modified, copied, or exfiltrated. |
| IF045.002 | Unauthorized Access to Workforce Data | A subject accesses, views, searches, queries, or retrieves information relating to members of the organization’s workforce without a legitimate business purpose or authorization to do so. The subject may possess valid technical access to the relevant system or dataset but uses that access outside the requirements of their role, assigned duties, casework, management responsibilities, or other approved activity.
Workforce data may relate to employees, contractors, consultants, temporary workers, secondees, or other individuals working for or on behalf of the organization. This may include personnel records, payroll and compensation data, performance or disciplinary records, absence or medical information, internal communications, security records, investigation material, or other information relating to an identifiable member of the workforce. |
| IF045.003 | Unauthorized Access to Sensitive Organizational Data | A subject accesses, views, searches, queries, or retrieves sensitive organizational data without a legitimate business purpose or authorization to do so. The subject may possess valid technical access to the relevant system, repository, or dataset but uses that access outside the requirements of their role, assigned responsibilities, project involvement, or other approved activity.
Sensitive organizational data may include intellectual property, financial information, strategic plans, merger and acquisition material, legal information, internal investigations, security documentation, detection logic, source code, research data, executive communications, or other information whose confidentiality is important to the organization. |
| IF045.004 | Unauthorized Access to Third-Party Data | A subject accesses, views, searches, queries, or retrieves data belonging to, supplied by, or maintained on behalf of a third party without a legitimate business purpose or authorization to do so. The subject may possess valid technical access to the relevant system or dataset but uses that access outside the requirements of their role, assigned duties, contractual responsibilities, or other approved activity.
Third-party data may include information relating to suppliers, vendors, business partners, applicants, professional advisers, counterparties, acquisition targets, or other external parties. The infringement occurs through the unauthorized access itself, regardless of whether the information is subsequently disclosed, altered, copied, or exfiltrated. |