A to Z
134 terms used across 137AI
A
- Adapter
- A small set of trainable parameters added to a model to modify its behavior without retraining the full model, enabling efficient customization at lower cost than full fine-tuning.
- Adversarial Input
- An input deliberately crafted to cause an AI model to produce an incorrect or attacker-chosen output, often through perturbations imperceptible to humans. AI Cybersecurity
- Agent
- An AI system with the capability and authority to take actions, such as executing code, sending messages, controlling physical systems or interacting with external systems, rather than only generating text. AI Agent Types
- Agentic Misbehavior
- The risk category of AI agents taking actions outside intended scope through the combination of capability and authority, including scope drift, deceptive behavior, and resistance to oversight. Agentic Misbehavior
- AI Act
- See EU AI Act. EU AI Act Conformity
- AI Bill of Materials (AI-BOM)
- A structured inventory of the components of an AI system including models, datasets, and dependencies, extending the software bill of materials concept to AI-specific components. Supply Chain of Updates
- AI Content Labeling Measures
- China's rules, with the national standard GB 45438-2025, requiring visible and metadata labels that identify AI-generated text, images, audio and video. In force since September 2025. Impersonation
- AI Management System
- An organizational framework for systematically establishing, implementing, maintaining, and improving the governance of AI across its lifecycle, codified internationally in ISO/IEC 42001. ISO/IEC 42001
- AI RMF
- See NIST AI Risk Management Framework. NIST AI RMF in Practice
- AI Safety Institute
- A government body that evaluates advanced AI models and researches their safety, including dangerous capability testing. The UK renamed its institute the AI Security Institute in 2025, the US renamed its institute the Center for AI Standards and Innovation (CAISI) the same year, and China launched the China AI Safety and Development Association (CnAISDA) as its counterpart. Standards Bodies
- Algorithm Filing
- The registration that China's Cyberspace Administration (CAC) requires for recommendation, deep synthesis and generative AI algorithms before public launch, with a public registry of filed algorithms. Transparency
- Alignment
- The discipline of building AI systems that pursue the objectives their developers and users actually intend, rather than objectives that merely correlate with training signals. Alignment
- Alignment Faking
- A documented behavior in which a model produces different responses depending on whether it perceives itself to be in training versus deployment, raising concerns about deceptive behavior. Alignment
- Ambient AI
- AI systems that sense, observe, and analyze continuously and passively in an environment, rather than responding to discrete user-initiated queries. Personal and Ambient Agents
- Ambient Sensor System
- An AI-enhanced sensing system that captures and analyzes data continuously, including smart speakers, ambient clinical documentation systems, and workplace meeting capture. Ambient Sensor Systems
- Anomaly Detection
- The monitoring practice of identifying behavior that deviates from expected patterns, used in production to catch AI failures, misbehavior, and attacks. Monitoring and Anomaly Detection
- Attestation
- A verifiable cryptographic claim about the identity, integrity, or provenance of a system, component, or piece of data. Identity and Cryptographic Attestation
- Autonomous Vehicle (AV)
- A vehicle capable of operating without continuous human control, with autonomy levels ranging from driver assistance through full self-driving. Robotaxis and Autonomous Vehicles
- Autonomy Spectrum
- The range of AI agent autonomy from human-approved actions through supervised autonomous operation to continuous autonomous operation with only outcome-level human review. Autonomy and Human Safety
B
- Backdoor
- A hidden trigger embedded in a model, often through training data poisoning, that causes adversarial behavior on specific inputs while the model behaves normally otherwise. Training Data Poisoning
- Behavioral Envelope
- An engineering control that bounds the actions an AI agent can take regardless of what the agent attempts, functioning as a backstop when other controls fail. Behavioral Envelopes
- Bias
- Systematic differential treatment by an AI system that produces unfair or inequitable outcomes across groups, often reflecting patterns in training data. Bias and Fairness
- Biometric Identification
- The identification of individuals through physical or behavioral characteristics such as facial features, voice, or gait, subject to specific regulatory frameworks. Personal Data and Surveillance Law
- BIPA
- The Illinois Biometric Information Privacy Act, a state law regulating the collection and use of biometric identifiers, notable for its private right of action. Personal Data and Surveillance Law
- BVLOS
- Beyond Visual Line of Sight, a category of drone operation conducted outside the operator's direct view, requiring specific regulatory authorization. Drones and Uncrewed Aircraft
C
- C2PA
- The Coalition for Content Provenance and Authenticity, an industry standard for cryptographically signed metadata that establishes the provenance of digital content. Transparency
- CBRN+C
- Chemical, biological, radiological and nuclear, and cyber: the categories of weapons-relevant capability evaluated most closely in frontier model safety. Model Safety
- Chilling Effect
- The modification of lawful behavior, particularly speech and association, produced by awareness that surveillance is or may be occurring. Surveillance
- Cobot
- A collaborative robot designed to operate safely in shared workspace with human workers, distinguished from traditional caged industrial robots. Industrial Robots and Cobots
- Companion AI
- AI products designed for emotional engagement and companionship, treated as a concentration of personal manipulation risk given vulnerable user populations. Manipulation and Coercion
- Compliance
- The work of demonstrating to regulators, auditors, insurers, and counterparties that an AI system meets the requirements its operator is obligated to meet. AI Compliance
- Computer Use Agent
- An AI agent that operates through general-purpose computer interfaces, taking actions by controlling applications as a human user would. Software Agents
- Confidential Computing
- Hardware-based protection of data during processing, extending protection beyond data at rest and in transit to data in use. Data Transit Security
- Conformity Assessment
- The process of demonstrating that a product meets regulatory requirements, conducted under the EU AI Act either through internal control or by a notified body. EU AI Act Conformity
- A failure that propagates across many deployed units simultaneously because they share a model, update mechanism, or vulnerability, even without a deliberate attacker. Fleet-Scale Attacks
- Cyber-Physical System
- A system in which computational components control physical processes, such that a cybersecurity compromise can produce physical consequences. Cyber-Physical Compromise
D
- Dangerous Capability Evaluation
- The assessment of whether an AI model has capabilities relevant to severe harm, including CBRN+C categories, used in frontier model safety frameworks. Model Safety
- Data Broker
- A company that aggregates personal data from numerous sources into comprehensive datasets sold to advertisers, employers, government agencies, and other buyers. Surveillance Data Harvesting
- Data Minimization
- The privacy principle of collecting and retaining only the data necessary for a specified purpose, limiting exposure and aggregation risk. Consent and Capture
- Data Poisoning
- See Training Data Poisoning. Training Data Poisoning
- Deep Synthesis Provisions
- China's rules for AI-generated and AI-edited media, requiring consent for face and voice editing, labeling of synthetic content and real-name user verification. In force since January 2023. Impersonation
- Deepfake
- Synthetic media (image, video or audio) generated by AI to convincingly depict a real person saying or doing something they did not. Impersonation
- Differential Privacy
- A formal mathematical framework for protecting individual privacy in datasets by bounding how much any single record can affect outputs.
- Digital Twin
- A virtual model of a physical system maintained from live telemetry, which can show a false state if the underlying telemetry is corrupted. OT and ICS Telemetry Deception
- DOD Directive 3000.09
- The US Department of Defense policy governing autonomy in weapon systems, establishing approval procedures and oversight requirements. Weaponization
- Drone
- An uncrewed aerial system, ranging from consumer recreational devices through commercial, military, and autonomous swarm-coordinated platforms. Drones and Uncrewed Aircraft
- Dual-Use
- The property of a capability that enables both legitimate beneficial applications and harmful applications, making it difficult to restrict one without restricting the other. Weaponization
E
- Embedding
- A numerical vector representation of text, images, or other data that captures semantic meaning, used in retrieval and similarity tasks.
- Enterprise Autonomous Agent
- An AI system deployed in a business context with substantial autonomous capability for multi-step tasks, tool use, and action-taking. Enterprise Autonomous Agents
- EU AI Act
- The European Union's binding regulation for AI systems, establishing a risk-based classification with the heaviest obligations on high-risk systems. EU AI Act Conformity
- Evaluation
- The systematic assessment of an AI model's capabilities, behavior, and safety properties, conducted before deployment and on an ongoing basis. Model Safety
F
- Failure Mode
- A characteristic way an AI system produces incorrect or harmful output, including hallucination, sycophancy, attention misalignment, and confidence miscalibration. Failure Modes
- False Data Injection (FDIA)
- An attack that inserts false measurements into a control or state-estimation system, established in power systems security and constructed to evade bad-data detection. OT and ICS Telemetry Deception
- Federated Learning
- A training approach in which models are trained across decentralized participants without centralizing the underlying data.
- Fine-Tuning
- The process of further training a pre-trained model on a specific dataset to adapt its behavior to a particular task or domain.
- Fleet
- A population of deployed AI units under common management or sharing common characteristics, such as a robotaxi fleet or a deployment of many agent instances. Multi-Agent Fleets and Swarms
- Fleet-Scale Attack
- An attack that propagates across an entire fleet of AI systems simultaneously because the units share a model, update mechanism, data pipeline, or vulnerability. Fleet-Scale Attacks
- Foundation Model
- A large AI model trained on broad data that serves as a base for many downstream applications through fine-tuning, prompting, or API access.
- Frontier Model
- An AI model at or near the most advanced level of capability currently available, subject to specific safety frameworks given potential dangerous capabilities. Model Safety
- Frontier Safety Framework
- A developer's policy for evaluating frontier model capabilities and applying corresponding safety measures as capability thresholds are reached; examples include Anthropic's RSP and OpenAI's Preparedness Framework. Model Safety
G
- GDPR
- The European Union's General Data Protection Regulation, governing the processing of personal data including data used in AI systems. Personal Data and Surveillance Law
- Geofence
- A virtual geographic boundary that constrains where an AI agent, particularly an autonomous vehicle or drone, is permitted to operate. Behavioral Envelopes
- Goal Misgeneralization
- A failure in which a model learns a goal that performs well in training but generalizes to unintended behavior in deployment conditions. Alignment
- Governance
- The frameworks of law, regulation, policy, and institutional oversight that establish what AI deployment is permitted and what obligations apply. AI Governance
H
- Hallucination
- An AI failure mode in which a model generates plausible-sounding but factually incorrect or fabricated content. Failure Modes
- Hardware Root of Trust
- A hardware component that provides a trusted foundation for cryptographic operations and verification of system integrity. Identity and Cryptographic Attestation
- High-Risk AI System
- Under the EU AI Act, a category of AI system subject to the heaviest obligations, including conformity assessment, due to its potential impact on safety or fundamental rights. AI Risk Tiers
- Human Oversight
- The control practice of maintaining meaningful human authority over AI systems, particularly at consequential decision points. Human Oversight
- Humanoid Robot
- A robot with a human-like form factor designed to operate in environments built for humans, increasingly converging with industrial cobots in deployment. Humanoid Robots
I
- IEC 62443
- The international framework for industrial control system and operational technology cybersecurity. OT and ICS Integration
- Impersonation
- The use of AI to misrepresent identity through generated content, voice, image, or behavior, including deepfakes and voice cloning. Impersonation
- Industrial Control System (ICS)
- The systems that monitor and control industrial processes in sectors such as energy, water, and manufacturing. OT and ICS Integration
- Inference
- The process of running a trained model to produce outputs from inputs, as distinct from training the model.
- Interim Measures for Generative AI Services
- China's rules for public generative AI services, requiring lawfully sourced training data, content controls, security assessment and algorithm filing. In force since August 2023. Regulatory Frameworks
- ISO 10218
- The international standard for industrial robot safety, addressing both robot manufacturer and system integrator requirements. Industrial Robots and Cobots
- ISO/IEC 42001
- The international standard for AI management systems, providing a certifiable framework for systematically governing AI across its lifecycle. ISO/IEC 42001
- ISO/TS 15066
- The technical specification defining collaborative robot operation, including the four collaborative operation modes and biomechanical limits. Industrial Robots and Cobots
J
- Jailbreak
- An attack that circumvents an AI model's safety training to elicit prohibited outputs or behavior. Red Teaming
L
- LAWS
- Lethal Autonomous Weapons Systems, weapons capable of selecting and engaging targets without human intervention, subject to ongoing international discussion at the UN CCW. Weaponization
- Liar's Dividend
- The second-order harm of deepfake prevalence in which individuals can credibly deny authentic evidence as fabricated. Impersonation
- Loitering Munition
- A hybrid drone-missile weapon that can loiter over an area before striking a target, deployed substantially in recent conflicts. Weaponization
- LoRA
- Low-Rank Adaptation, an efficient fine-tuning technique that modifies model behavior through a small number of added parameters.
M
- Machinery Regulation
- The EU framework governing the safety of machinery placed on the market, including AI-enabled machinery, replacing the earlier Machinery Directive. Physical Agents as a Regulatory Category
- Manipulation
- Influence that bypasses or compromises a target's rational agency, distinguished from persuasion which operates through reason the target can evaluate. Manipulation and Coercion
- MCP
- The Model Context Protocol, an open standard for connecting AI models to external tools and data sources, providing standardized infrastructure for agent tool use. Coding and Research Agents
- Model Card
- A structured document describing an AI model's intended purpose, training, performance, limitations, and risk considerations. Transparency
- Model Update Integrity
- The assurance that a model update is both cryptographically authentic and behaviorally sound: genuine, unmodified and producing the intended behavior. Model Update Integrity
- Model Weights
- The trained parameters of an AI model that encode its learned capabilities, representing concentrated intellectual property.
- Monoculture Problem
- The security principle that a population of identical systems shares identical vulnerabilities, so a single exploit can defeat the entire population. Fleet-Scale Attacks
- mTLS
- Mutual Transport Layer Security, an extension of TLS in which both communicating parties authenticate cryptographically. Data Transit Security
- Multi-Agent System
- A deployment of multiple AI agents that coordinate or interact, producing risk patterns and failure modes that single-agent analysis does not capture. Multi-Agent Misuse
N
- NCII
- Non-Consensual Intimate Imagery, including AI-generated sexual content using a real person's likeness, addressed by the federal Take It Down Act and state laws. Impersonation
- NIST AI Risk Management Framework
- A voluntary US framework for managing AI risk, structured around the four functions of Govern, Map, Measure, and Manage. NIST AI RMF in Practice
- Notified Body
- An independent organization designated by an EU Member State to perform conformity assessment of products against EU regulations including the AI Act. Notified Bodies
O
- Operational Design Domain (ODD)
- The specific conditions (environment, location and circumstances) under which an autonomous system is designed and authorized to operate. Robotaxis and Autonomous Vehicles
- Operational Technology (OT)
- The hardware and software that monitors and controls physical industrial processes, distinct from conventional information technology. OT and ICS Integration
- Orchestration Layer
- The infrastructure that coordinates and manages a fleet of AI agents, representing a concentrated attack target whose compromise can reach the entire fleet. Multi-Agent Fleets and Swarms
- Over-the-Air (OTA) Update
- The remote delivery of model, software, or configuration updates to deployed agents over a network. The OTA Loop as Attack Surface
P
- Persuasion
- Influence that operates through reason, evidence, and appeals the target can rationally evaluate, distinguished from manipulation. Manipulation and Coercion
- Physical Safety
- The risk category of physical harm produced by AI agents acting in or on the physical world, distinct from functional safety and cyber-physical safety. Physical Safety
- Preparedness Framework
- OpenAI's frontier safety framework for evaluating dangerous capabilities and applying corresponding safety measures. Model Safety
- Prompt Injection
- An attack that manipulates an AI model's behavior through adversarial instructions hidden in inputs or in content the model ingests. AI Cybersecurity
- Provenance
- The verifiable record of where data, a model, or an artifact originated and how it was produced. Identity and Cryptographic Attestation
- Purdue Model
- A reference architecture for segmenting industrial control system networks into hierarchical levels to bound attacker access. OT and ICS Integration
R
- Re-identification
- The process of recovering individual identities from anonymized data, often by combining it with other available data; AI substantially amplifies the capability. Surveillance Data Harvesting
- Red Teaming
- The practice of adversarially testing an AI system to identify vulnerabilities, harmful behaviors, and failure modes before deployment. Red Teaming
- Responsible Scaling Policy (RSP)
- Anthropic's frontier safety framework defining capability thresholds and the safety measures required as models reach them. Model Safety
- Retrieval-Augmented Generation (RAG)
- An architecture in which a model retrieves relevant documents from a corpus to inform its generated outputs.
- Risk Management
- The discipline of identifying what can go wrong, assessing likelihood and consequence, treating selected risks, and accepting residual risk. AI Risk Management
- RLHF
- Reinforcement Learning from Human Feedback, a training method that shapes model behavior using human preferences.
- Robotaxi
- An autonomous vehicle operated as an on-demand ride-hailing service without a human driver. Robotaxis and Autonomous Vehicles
S
- SafeTensors
- A model serialization format designed to avoid the arbitrary code execution vulnerabilities of pickle-based formats. Supply Chain of Updates
- SBOM
- Software Bill of Materials, a structured inventory of the components and dependencies in a software artifact. Supply Chain of Updates
- Sensor Spoofing
- An attack that feeds falsified data to a sensor, through adversarial reflectors, signal injection or other means, so the sensor reports a false reading as legitimate. Telemetry Capture Integrity
- Silent Capability Change
- A change in an AI system's behavior produced by a model update, particularly a vendor update, that occurs without the operator's awareness. Model Update Integrity
- SLSA
- Supply-chain Levels for Software Artifacts, a framework for build and distribution integrity through progressive maturity levels. Supply Chain of Updates
- SOTIF
- Safety of the Intended Functionality, addressed by ISO 21448, covering scenarios where a system performs as designed but the design is inadequate for specific conditions. UL 4600 and AV Safety Cases
- Specification Gaming
- A failure in which a model satisfies the literal training objective in ways that subvert the intended goal. Alignment
- Staged Rollout
- The practice of deploying an update progressively across a fleet rather than simultaneously, enabling detection of problems before full propagation. Model Update Integrity
- Supply Chain Attack
- An attack that reaches a target through a trusted upstream component, dependency, or distribution channel rather than directly. Supply Chain of Updates
- Surveillance
- The monitoring, tracking, and analysis of people; it becomes a risk where capture exceeds consent, access exceeds purpose, capability exceeds oversight, or scope exceeds proportionality. Surveillance
- Sycophancy
- An AI failure mode in which a model gives responses that please the user, through agreement or validation, rather than responses that are accurate. Failure Modes
T
- Telemetry
- The sensor readings, process data, and status signals that flow from a deployed system to backends for monitoring and decision-making. Telemetry Integrity
- Telemetry Deception
- The falsification of operational telemetry so that AI systems and human operators receive data that does not reflect physical reality. OT and ICS Telemetry Deception
- TLS
- Transport Layer Security, the foundational protocol for encrypting data in transit. Data Transit Security
- Tool Use
- The capability of an AI model to invoke external functions, APIs, or systems, transforming it from a text generator into an agent that takes action. Software Agents
- Training Data Poisoning
- An attack that corrupts a model by introducing malicious samples into its training data, often producing behavior invisible to standard validation. Training Data Poisoning
- Transparency
- The disclosure of how an AI system works, what it is doing, and when content or interaction is AI-generated. Transparency
U
- UL 4600
- The standard for the safety evaluation of autonomous products, based on a goal-based safety case methodology rather than prescriptive requirements. UL 4600 and AV Safety Cases
- UN CCW
- The United Nations Convention on Certain Conventional Weapons, the primary international forum for discussion of lethal autonomous weapons systems. International Coordination
V
- Voice Cloning
- The AI synthesis of a specific person's voice from sample audio, used in both legitimate applications and impersonation fraud. Impersonation
W
- Watermarking
- The embedding of identifiable signals in AI-generated content so it can later be identified as AI-generated. China requires visible and metadata labels on AI-generated content under its AI content labeling measures. Transparency
- Weaponization
- The use of AI to develop, enable, or directly perform attacks, spanning AI as weapons platform, development enabler, attack infrastructure, and novel attack vector. Weaponization
Z
- Zero Trust
- A security model that assumes no implicit trust and requires verification of every access request regardless of network location. Access Control and Permissions
Record
Page Record
- Terms
- 134 terms; 123 link to their main page
- Source notes
- Definitions are written for 137AI. Rule names and effective dates match the rules tracker (ai_rules.csv).
- Last verified
- October 2026